本篇介绍
我们在使用binder时候的大致流程是先获取servicemanger的binder，然后通过该binder获取目标服务的binder，最后调用该binder的接口。本篇就介绍下第一步的内容，涉及了从native到驱动，fwk部分先不涉及，希望可以完全理解servicemanager的获取流程。
内容介绍
整个流程的起点是IServiceManager::defaultServiceManager()
代码语言：javascript复制sp<IServiceManager> defaultServiceManager()
{
std::call_once(gSmOnce, []() {
sp<AidlServiceManager> sm = nullptr;
while (sm == nullptr) {
sm = interface_cast<AidlServiceManager>(ProcessState::self()->getContextObject(nullptr));//获取servicemanager的binder
if (sm == nullptr) {
ALOGE("Waiting 1s on context object on %s.", ProcessState::self()->getDriverName().c_str());
sleep(1);
}
}
gDefaultServiceManager = sp<ServiceManagerShim>::make(sm);
});
return gDefaultServiceManager;
}可以看到这是一个单例，在首次调用的时候会进行初始化获取servicemanager的binder，然后用智能指针封装一下。
代码语言：javascript复制std::call_once 这个是c 11上的一个新特性，用于需要保证某个流程在多线程场景中也只执行一次，需要和std::once_flag配套使用，用来区分不同的流程。 接下来看下ProcessState::self()->getContextObject(nullptr) 的内容:
#ifdef __ANDROID_VNDK__
const char* kDefaultDriver = "/dev/vndbinder";
#else
const char* kDefaultDriver = "/dev/binder";
#endif
sp<ProcessState> ProcessState::self()
{
return init(kDefaultDriver, false /*requireDefault*/);
}
sp<ProcessState> ProcessState::init(const char *driver, bool requireDefault)
{
[[clang::no_destroy]] static sp<ProcessState> gProcess;
[[clang::no_destroy]] static std::mutex gProcessMutex;
if (driver == nullptr) {
std::lock_guard<std::mutex> l(gProcessMutex);
return gProcess;
}
[[clang::no_destroy]] static std::once_flag gProcessOnce;
std::call_once(gProcessOnce, [&](){
if (access(driver, R_OK) == -1) {
ALOGE("Binder driver %s is unavailable. Using /dev/binder instead.", driver);
driver = "/dev/binder";
}
std::lock_guard<std::mutex> l(gProcessMutex);
gProcess = sp<ProcessState>::make(driver); // 创建一个processstate
});
if (requireDefault) {
// Detect if we are trying to initialize with a different driver, and
// consider that an error. ProcessState will only be initialized once above.
LOG_ALWAYS_FATAL_IF(gProcess->getDriverName() != driver,
"ProcessState was already initialized with %s,"
" can't initialize with %s.",
gProcess->getDriverName().c_str(), driver);
}
return gProcess;
}可以看到这儿还是一个单例，对于参与binder IPC的进程，和binder驱动交互部分就是通过ProcessState实现的。 对于android，有三种binder驱动节点，下面列一下区别：
binder驱动节点 | 描述 |
|---|---|
/dev/binder | 供android fwk使用aidl接口时使用 |
/dev/hwbinder | 供使用hidl接口的进程使用，主要是fwk和vendor进程，vendor进程之间 |
/dev/vndbinder | 供使用aidl接口的vendor进程之间使用 |
接下来看下ProcessState的构造方法：
代码语言：javascript复制ProcessState::ProcessState(const char *driver)
: mDriverName(String8(driver))
, mDriverFD(open_driver(driver)) // 打开 binder节点
, mVMStart(MAP_FAILED)
, mThreadCountLock(PTHREAD_MUTEX_INITIALIZER)
, mThreadCountDecrement(PTHREAD_COND_INITIALIZER)
, mExecutingThreadsCount(0)
, mWaitingForThreads(0)
, mMaxThreads(DEFAULT_MAX_BINDER_THREADS) // 线程池数量
, mStarvationStartTimeMs(0)
, mThreadPoolStarted(false)
, mThreadPoolSeq(1)
, mCallRestriction(CallRestriction::NONE)
{
// TODO(b/166468760): enforce in build system
#if defined(__ANDROID_APEX__)
LOG_ALWAYS_FATAL("Cannot use libbinder in APEX (only system.img libbinder) since it is not stable.");
#endif
if (mDriverFD >= 0) {
// mmap the binder, providing a chunk of virtual address space to receive transactions.
mVMStart = mmap(nullptr, BINDER_VM_SIZE, PROT_READ, MAP_PRIVATE | MAP_NORESERVE, mDriverFD, 0); // 内存映射
if (mVMStart == MAP_FAILED) {
// *sigh*
ALOGE("Using %s failed: unable to mmap transaction memory.n", mDriverName.c_str());
close(mDriverFD);
mDriverFD = -1;
mDriverName.clear();
}
}
#ifdef __ANDROID__
LOG_ALWAYS_FATAL_IF(mDriverFD < 0, "Binder driver '%s' could not be opened. Terminating.", driver);
#endif
}这儿有两个和驱动交互的部分，一个是open_driver，一个是mmap，这里先介绍下open_driver：
代码语言：javascript复制static int open_driver(const char *driver)
{
int fd = open(driver, O_RDWR | O_CLOEXEC); //打开binder节点
if (fd >= 0) {
int vers = 0;
status_t result = ioctl(fd, BINDER_VERSION, &vers); //获取binder版本号
if (result == -1) {
ALOGE("Binder ioctl to obtain version failed: %s", strerror(errno));
close(fd);
fd = -1;
}
if (result != 0 || vers != BINDER_CURRENT_PROTOCOL_VERSION) {
ALOGE("Binder driver protocol(%d) does not match user space protocol(%d)! ioctl() return value: %d",
vers, BINDER_CURRENT_PROTOCOL_VERSION, result);
close(fd);
fd = -1;
}
size_t maxThreads = DEFAULT_MAX_BINDER_THREADS;
result = ioctl(fd, BINDER_SET_MAX_THREADS, &maxThreads); //设置线程池数量
if (result == -1) {
ALOGE("Binder ioctl to set max threads failed: %s", strerror(errno));
}
uint32_t enable = DEFAULT_ENABLE_ONEWAY_SPAM_DETECTION;
result = ioctl(fd, BINDER_ENABLE_ONEWAY_SPAM_DETECTION, &enable);
if (result == -1) {
ALOGI("Binder ioctl to enable oneway spam detection failed: %s", strerror(errno));
}
} else {
ALOGW("Opening '%s' failed: %sn", driver, strerror(errno));
}
return fd;
}这儿的流程比较直接，就是打开节点，用ioctl获取了版本号，设置了线程池数量，设置了允许异步调用检测。接下来看下open 的内容，这时候就会进入内核，因为binder驱动定义了自己的open，ioctl，mmap方法。
代码语言：javascript复制static int __init binder_init(void)
{
int ret;
char *device_name, *device_tmp;
struct binder_device *device;
struct hlist_node *tmp;
char *device_names = NULL;
ret = binder_alloc_shrinker_init(); //cache管理初始化
if (ret)
return ret;
atomic_set(&binder_transaction_log.cur, ~0U);
atomic_set(&binder_transaction_log_failed.cur, ~0U);
binder_debugfs_dir_entry_root = debugfs_create_dir("binder", NULL); //debugfs目录，在android里面是/sys/kernel/debug，通常还有一个软连接/d 指向该路径
if (binder_debugfs_dir_entry_root)
binder_debugfs_dir_entry_proc = debugfs_create_dir("proc",
binder_debugfs_dir_entry_root);
if (binder_debugfs_dir_entry_root) {
debugfs_create_file("state",
0444,
binder_debugfs_dir_entry_root,
NULL,
&binder_state_fops);
debugfs_create_file("stats",
0444,
binder_debugfs_dir_entry_root,
NULL,
&binder_stats_fops);
debugfs_create_file("transactions",
0444,
binder_debugfs_dir_entry_root,
NULL,
&binder_transactions_fops);
debugfs_create_file("transaction_log",
0444,
binder_debugfs_dir_entry_root,
&binder_transaction_log,
&binder_transaction_log_fops);
debugfs_create_file("failed_transaction_log",
0444,
binder_debugfs_dir_entry_root,
&binder_transaction_log_failed,
&binder_transaction_log_fops);
}
if (!IS_ENABLED(CONFIG_ANDROID_BINDERFS) &&
strcmp(binder_devices_param, "") != 0) {
/*
* Copy the module_parameter string, because we don't want to
* tokenize it in-place.
*/
device_names = kstrdup(binder_devices_param, GFP_KERNEL);
if (!device_names) {
ret = -ENOMEM;
goto err_alloc_device_names_failed;
}
device_tmp = device_names;
while ((device_name = strsep(&device_tmp, ","))) {
ret = init_binder_device(device_name); // 初始化binder驱动设备，这儿的device_name 就包含了/dev/binder，/dev/hwbinder， /dev/vndbinder
if (ret)
goto err_init_binder_device_failed;
}
}
ret = init_binderfs();
if (ret)
goto err_init_binder_device_failed;
return ret;
err_init_binder_device_failed:
hlist_for_each_entry_safe(device, tmp, &binder_devices, hlist) {
misc_deregister(&device->miscdev);
hlist_del(&device->hlist);
kfree(device);
}
kfree(device_names);
err_alloc_device_names_failed:
debugfs_remove_recursive(binder_debugfs_dir_entry_root);
return ret;
}
device_initcall(binder_init);接下来看下init_binder_device的实现：
代码语言：javascript复制static int __init init_binder_device(const char *name)
{
int ret;
struct binder_device *binder_device;
binder_device = kzalloc(sizeof(*binder_device), GFP_KERNEL);
if (!binder_device)
return -ENOMEM;
binder_device->miscdev.fops = &binder_fops;
binder_device->miscdev.minor = MISC_DYNAMIC_MINOR;
binder_device->miscdev.name = name;
refcount_set(&binder_device->ref, 1);
binder_device->context.binder_context_mgr_uid = INVALID_UID;
binder_device->context.name = name;
mutex_init(&binder_device->context.context_mgr_node_lock);
ret = misc_register(&binder_device->miscdev); //将binder 驱动注册成为misc设备
if (ret < 0) {
kfree(binder_device);
return ret;
}
hlist_add_head(&binder_device->hlist, &binder_devices);
return ret;
}
const struct file_operations binder_fops = {
.owner = THIS_MODULE,
.poll = binder_poll,
.unlocked_ioctl = binder_ioctl,
.compat_ioctl = compat_ptr_ioctl,
.mmap = binder_mmap,
.open = binder_open,
.flush = binder_flush,
.release = binder_release,
};这儿就是将binder驱动设备注册为一个misc设备，并指定了它的操作实现方法。 因为上层是先调用的open，因此看下这儿binder_open的实现：
代码语言：javascript复制static int binder_open(struct inode *nodp, struct file *filp)
{
struct binder_proc *proc, *itr;
struct binder_device *binder_dev;
struct binderfs_info *info;
struct dentry *binder_binderfs_dir_entry_proc = NULL;
bool existing_pid = false;
binder_debug(BINDER_DEBUG_OPEN_CLOSE, "%s: %d:%dn", __func__,
current->group_leader->pid, current->pid);
proc = kzalloc(sizeof(*proc), GFP_KERNEL); // 一个binder进程对应一个binder_proc
if (proc == NULL)
return -ENOMEM;
spin_lock_init(&proc->inner_lock);
spin_lock_init(&proc->outer_lock);
get_task_struct(current->group_leader);
proc->tsk = current->group_leader;
INIT_LIST_HEAD(&proc->todo);
proc->default_priority = task_nice(current);
/* binderfs stashes devices in i_private */
if (is_binderfs_device(nodp)) {
binder_dev = nodp->i_private;
info = nodp->i_sb->s_fs_info;
binder_binderfs_dir_entry_proc = info->proc_log_dir;
} else {
binder_dev = container_of(filp->private_data,
struct binder_device, miscdev);
}
refcount_inc(&binder_dev->ref);
proc->context = &binder_dev->context;
binder_alloc_init(&proc->alloc);
binder_stats_created(BINDER_STAT_PROC);
proc->pid = current->group_leader->pid;
INIT_LIST_HEAD(&proc->delivered_death);
INIT_LIST_HEAD(&proc->waiting_threads);
filp->private_data = proc;
mutex_lock(&binder_procs_lock);
hlist_for_each_entry(itr, &binder_procs, proc_node) { // 加入binder_proc链表
if (itr->pid == proc->pid) {
existing_pid = true;
break;
}
}
hlist_add_head(&proc->proc_node, &binder_procs);
mutex_unlock(&binder_procs_lock);
if (binder_debugfs_dir_entry_proc && !existing_pid) {
char strbuf[11];
snprintf(strbuf, sizeof(strbuf), "%u", proc->pid);
/*
* proc debug entries are shared between contexts.
* Only create for the first PID to avoid debugfs log spamming
* The printing code will anyway print all contexts for a given
* PID so this is not a problem.
*/
proc->debugfs_entry = debugfs_create_file(strbuf, 0444,
binder_debugfs_dir_entry_proc,
(void *)(unsigned long)proc->pid,
&proc_fops);
}
if (binder_binderfs_dir_entry_proc && !existing_pid) {
char strbuf[11];
struct dentry *binderfs_entry;
snprintf(strbuf, sizeof(strbuf), "%u", proc->pid);
/*
* Similar to debugfs, the process specific log file is shared
* between contexts. Only create for the first PID.
* This is ok since same as debugfs, the log file will contain
* information on all contexts of a given PID.
*/
binderfs_entry = binderfs_create_file(binder_binderfs_dir_entry_proc,
strbuf, &proc_fops, (void *)(unsigned long)proc->pid);
if (!IS_ERR(binderfs_entry)) {
proc->binderfs_entry = binderfs_entry;
} else {
int error;
error = PTR_ERR(binderfs_entry);
pr_warn("Unable to create file %s in binderfs (error %d)n",
strbuf, error);
}
}
return 0;
}这儿主要就是创建了一个binder_proc结构，并和当前用户态进程current关联起来。 再看下ioctl的内容：
代码语言：javascript复制static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
int ret;
struct binder_proc *proc = filp->private_data;
struct binder_thread *thread;
unsigned int size = _IOC_SIZE(cmd);
void __user *ubuf = (void __user *)arg;
/*pr_info("binder_ioctl: %d:%d %x %lxn",
proc->pid, current->pid, cmd, arg);*/
binder_selftest_alloc(&proc->alloc);
trace_binder_ioctl(cmd, arg);
ret = wait_event_interruptible(binder_user_error_wait, binder_stop_on_user_error < 2);
if (ret)
goto err_unlocked;
thread = binder_get_thread(proc); // 获取一个binder线程
if (thread == NULL) {
ret = -ENOMEM;
goto err;
}
switch (cmd) {
case BINDER_WRITE_READ: // 用的最多的就是这个命令了
ret = binder_ioctl_write_read(filp, cmd, arg, thread);
if (ret)
goto err;
break;
case BINDER_SET_MAX_THREADS: { // 设置binder线程池数量
int max_threads;
if (copy_from_user(&max_threads, ubuf,
sizeof(max_threads))) {
ret = -EINVAL;
goto err;
}
binder_inner_proc_lock(proc);
proc->max_threads = max_threads; //在这儿设置生效
binder_inner_proc_unlock(proc);
break;
}
case BINDER_SET_CONTEXT_MGR_EXT: { //设置binder管家，也就是并不一定servicemanager就是binder管家，也可以将其他服务设置成binder管家
struct flat_binder_object fbo;
if (copy_from_user(&fbo, ubuf, sizeof(fbo))) {
ret = -EINVAL;
goto err;
}
ret = binder_ioctl_set_ctx_mgr(filp, &fbo);
if (ret)
goto err;
break;
}
case BINDER_SET_CONTEXT_MGR: //设置binder管家
ret = binder_ioctl_set_ctx_mgr(filp, NULL);
if (ret)
goto err;
break;
case BINDER_THREAD_EXIT:
binder_debug(BINDER_DEBUG_THREADS, "%d:%d exitn",
proc->pid, thread->pid);
binder_thread_release(proc, thread);
thread = NULL;
break;
case BINDER_VERSION: { // 获取binder版本
struct binder_version __user *ver = ubuf;
if (size != sizeof(struct binder_version)) {
ret = -EINVAL;
goto err;
}
if (put_user(BINDER_CURRENT_PROTOCOL_VERSION,
&ver->protocol_version)) {
ret = -EINVAL;
goto err;
}
break;
}
case BINDER_GET_NODE_INFO_FOR_REF: {
struct binder_node_info_for_ref info;
if (copy_from_user(&info, ubuf, sizeof(info))) {
ret = -EFAULT;
goto err;
}
ret = binder_ioctl_get_node_info_for_ref(proc, &info);
if (ret < 0)
goto err;
if (copy_to_user(ubuf, &info, sizeof(info))) {
ret = -EFAULT;
goto err;
}
break;
}
case BINDER_GET_NODE_DEBUG_INFO: {
struct binder_node_debug_info info;
if (copy_from_user(&info, ubuf, sizeof(info))) {
ret = -EFAULT;
goto err;
}
ret = binder_ioctl_get_node_debug_info(proc, &info);
if (ret < 0)
goto err;
if (copy_to_user(ubuf, &info, sizeof(info))) {
ret = -EFAULT;
goto err;
}
break;
}
default:
ret = -EINVAL;
goto err;
}
ret = 0;
err:
if (thread)
thread->looper_need_return = false;
wait_event_interruptible(binder_user_error_wait, binder_stop_on_user_error < 2);
if (ret && ret != -ERESTARTSYS)
pr_info("%d:%d ioctl %x %lx returned %dn", proc->pid, current->pid, cmd, arg, ret);
err_unlocked:
trace_binder_ioctl_done(ret);
return ret;
}看到这一块，之前的三个ioctl命令内容就清楚了。接下来需要看下一个关键的地方，就是mmap，对应的实现就是binder_mmap:
代码语言：javascript复制static int binder_mmap(struct file *filp, struct vm_area_struct *vma)
{
struct binder_proc *proc = filp->private_data;
if (proc->tsk != current->group_leader)
return -EINVAL;
...
vma->vm_flags |= VM_DONTCOPY | VM_MIXEDMAP;
vma->vm_flags &= ~VM_MAYWRITE;
vma->vm_ops = &binder_vm_ops;
vma->vm_private_data = proc;
return binder_alloc_mmap_handler(&proc->alloc, vma); 这儿没做啥，看下binder_alloc_mmap_handler：
代码语言：javascript复制int binder_alloc_mmap_handler(struct binder_alloc *alloc,
struct vm_area_struct *vma)
{
int ret;
const char *failure_string;
struct binder_buffer *buffer;
mutex_lock(&binder_alloc_mmap_lock);
if (alloc->buffer_size) {
ret = -EBUSY;
failure_string = "already mapped";
goto err_already_mapped;
}
alloc->buffer_size = min_t(unsigned long, vma->vm_end - vma->vm_start,
SZ_4M); // 更新binder的buffer 空间大小
mutex_unlock(&binder_alloc_mmap_lock);
alloc->buffer = (void __user *)vma->vm_start; // 设置起始地址
alloc->pages = kcalloc(alloc->buffer_size / PAGE_SIZE,
sizeof(alloc->pages[0]),
GFP_KERNEL); // 分配页结构，只是页结构，并未真正分配物理内存
if (alloc->pages == NULL) {
ret = -ENOMEM;
failure_string = "alloc page array";
goto err_alloc_pages_failed;
}
buffer = kzalloc(sizeof(*buffer), GFP_KERNEL); // 分配buffer结构
if (!buffer) {
ret = -ENOMEM;
failure_string = "alloc buffer struct";
goto err_alloc_buf_struct_failed;
}
buffer->user_data = alloc->buffer;
list_add(&buffer->entry, &alloc->buffers); // 添加到buffer链表
buffer->free = 1;
binder_insert_free_buffer(alloc, buffer); // 添加到空间buffer链表
alloc->free_async_space = alloc->buffer_size / 2; // 更新异步调用buffer空间
binder_alloc_set_vma(alloc, vma); // 保存vma到alloc
mmgrab(alloc->vma_vm_mm);
return 0;
}这时候驱动做的事情就是将用户态的地址保存到proc自己的结构里面。 mmap暂时告一段落。 继续回到用户态，继续ProcessState::self()->getContextObject(nullptr) 获取serviemanager的binder过程，前一部分介绍完了，接下来应该是getContextObject部分，看下实现：
代码语言：javascript复制sp<IBinder> ProcessState::getContextObject(const sp<IBinder>& /*caller*/)
{
sp<IBinder> context = getStrongProxyForHandle(0);
if (context) {
// The root object is special since we get it directly from the driver, it is never
// written by Parcell::writeStrongBinder.
internal::Stability::markCompilationUnit(context.get());
} else {
ALOGW("Not able to get context object on %s.", mDriverName.c_str());
}
return context;
}对于servicemanager，handle为0，看下getStrongProxyForHandle的内容：
代码语言：javascript复制sp<IBinder> ProcessState::getStrongProxyForHandle(int32_t handle)
{
sp<IBinder> result;
AutoMutex _l(mLock);
handle_entry* e = lookupHandleLocked(handle); // 本地查找，刚开始肯定没有，因此handler_entry->binder 是nullptr
if (e != nullptr) {
// We need to create a new BpBinder if there isn't currently one, OR we
// are unable to acquire a weak reference on this current one. The
// attemptIncWeak() is safe because we know the BpBinder destructor will always
// call expungeHandle(), which acquires the same lock we are holding now.
// We need to do this because there is a race condition between someone
// releasing a reference on this BpBinder, and a new reference on its handle
// arriving from the driver.
IBinder* b = e->binder;
if (b == nullptr || !e->refs->attemptIncWeak(this)) {
if (handle == 0) {
// Special case for context manager...
// The context manager is the only object for which we create
// a BpBinder proxy without already holding a reference.
// Perform a dummy transaction to ensure the context manager
// is registered before we create the first local reference
// to it (which will occur when creating the BpBinder).
// If a local reference is created for the BpBinder when the
// context manager is not present, the driver will fail to
// provide a reference to the context manager, but the
// driver API does not return status.
//
// Note that this is not race-free if the context manager
// dies while this code runs.
//
// TODO: add a driver API to wait for context manager, or
// stop special casing handle 0 for context manager and add
// a driver API to get a handle to the context manager with
// proper reference counting.
IPCThreadState* ipc = IPCThreadState::self();
CallRestriction originalCallRestriction = ipc->getCallRestriction();
ipc->setCallRestriction(CallRestriction::NONE);
Parcel data;
status_t status = ipc->transact(
0, IBinder::PING_TRANSACTION, data, nullptr, 0);
ipc->setCallRestriction(originalCallRestriction);
if (status == DEAD_OBJECT)
return nullptr;
}
sp<BpBinder> b = BpBinder::create(handle);
e->binder = b.get();
if (b) e->refs = b->getWeakRefs();
result = b;
} else {
// This little bit of nastyness is to allow us to add a primary
// reference to the remote proxy when this team doesn't have one
// but another team is sending the handle to us.
result.force_set(b);
e->refs->decWeak(this);
}
}
return result;
}如果本地没有servicemanager的proxy binder，那么就需要用驱动获取。继续看下获取流程：
代码语言：javascript复制IPCThreadState* IPCThreadState::self()
{
if (gHaveTLS.load(std::memory_order_acquire)) {
restart:
const pthread_key_t k = gTLS;
IPCThreadState* st = (IPCThreadState*)pthread_getspecific(k); // 线程局部变量 tls
if (st) return st;
return new IPCThreadState;
}
...
}
IPCThreadState::IPCThreadState()
: mProcess(ProcessState::self()),
mServingStackPointer(nullptr),
mWorkSource(kUnsetWorkSource),
mPropagateWorkSource(false),
mIsLooper(false),
mStrictModePolicy(0),
mLastTransactionBinderFlags(0),
mCallRestriction(mProcess->mCallRestriction)
{
pthread_setspecific(gTLS, this); // 设置成线程局部变量，也就是每个线程一个ipcthreadstate
clearCaller();
mIn.setDataCapacity(256);
mOut.setDataCapacity(256);
}看下transact：
代码语言：javascript复制status_t IPCThreadState::transact(int32_t handle,
uint32_t code, const Parcel& data,
Parcel* reply, uint32_t flags)
{
status_t err;
flags |= TF_ACCEPT_FDS;
err = writeTransactionData(BC_TRANSACTION, flags, handle, code, data, nullptr);// 封装数据包
if (err != NO_ERROR) {
if (reply) reply->setError(err);
return (mLastError = err);
}
...
if (reply) {
err = waitForResponse(reply); // 和驱动交互
} else {
Parcel fakeReply;
err = waitForResponse(&fakeReply);
}
...
} else {
err = waitForResponse(nullptr, nullptr);
}
return err;
}可以看到transact里面的流程就是封装数据包和与驱动交互，如果有返回值，则直接写入reply。先看下如何封装的数据包：
代码语言：javascript复制status_t IPCThreadState::writeTransactionData(int32_t cmd, uint32_t binderFlags,
int32_t handle, uint32_t code, const Parcel& data, status_t* statusBuffer)
{
binder_transaction_data tr;
tr.target.ptr = 0; /* Don't pass uninitialized stack data to a remote process */
tr.target.handle = handle; // servicemanager 的handle， 也就是0
tr.code = code; // PING_TRANSACTION
tr.flags = binderFlags;
tr.cookie = 0;
tr.sender_pid = 0;
tr.sender_euid = 0;
const status_t err = data.errorCheck();
if (err == NO_ERROR) {
tr.data_size = data.ipcDataSize(); // 这时候应该都是空的
tr.data.ptr.buffer = data.ipcData();
tr.offsets_size = data.ipcObjectsCount()*sizeof(binder_size_t);
tr.data.ptr.offsets = data.ipcObjects();
} else if (statusBuffer) {
tr.flags |= TF_STATUS_CODE;
*statusBuffer = err;
tr.data_size = sizeof(status_t);
tr.data.ptr.buffer = reinterpret_cast<uintptr_t>(statusBuffer);
tr.offsets_size = 0;
tr.data.ptr.offsets = 0;
} else {
return (mLastError = err);
}
mOut.writeInt32(cmd); // BC_TRANSACTION
mOut.write(&tr, sizeof(tr)); // 将数据写入parcel
return NO_ERROR;
}这时候就把请求相关的写入了parcel，接下来看下waitForResponse：
代码语言：javascript复制status_t IPCThreadState::waitForResponse(Parcel *reply, status_t *acquireResult)
{
uint32_t cmd;
int32_t err;
while (1) {
if ((err=talkWithDriver()) < NO_ERROR) break; // 将请求写入内核，并读取驱动回复
err = mIn.errorCheck();
if (err < NO_ERROR) break;
if (mIn.dataAvail() == 0) continue;
cmd = (uint32_t)mIn.readInt32();
...
switch (cmd) { // 解析驱动的回复
case BR_ONEWAY_SPAM_SUSPECT:
ALOGE("Process seems to be sending too many oneway calls.");
CallStack::logStack("oneway spamming", CallStack::getCurrent().get(),
ANDROID_LOG_ERROR);
[[fallthrough]];
case BR_TRANSACTION_COMPLETE:
if (!reply && !acquireResult) goto finish;
break;
case BR_DEAD_REPLY:
err = DEAD_OBJECT;
goto finish;
case BR_FAILED_REPLY:
err = FAILED_TRANSACTION;
goto finish;
case BR_FROZEN_REPLY:
err = FAILED_TRANSACTION;
goto finish;
case BR_ACQUIRE_RESULT:
{
ALOG_ASSERT(acquireResult != NULL, "Unexpected brACQUIRE_RESULT");
const int32_t result = mIn.readInt32();
if (!acquireResult) continue;
*acquireResult = result ? NO_ERROR : INVALID_OPERATION;
}
goto finish;
case BR_REPLY:
{
binder_transaction_data tr;
err = mIn.read(&tr, sizeof(tr));
ALOG_ASSERT(err == NO_ERROR, "Not enough command data for brREPLY");
if (err != NO_ERROR) goto finish;
if (reply) {
if ((tr.flags & TF_STATUS_CODE) == 0) {
reply->ipcSetDataReference(
reinterpret_cast<const uint8_t*>(tr.data.ptr.buffer),
tr.data_size,
reinterpret_cast<const binder_size_t*>(tr.data.ptr.offsets),
tr.offsets_size/sizeof(binder_size_t),
freeBuffer);
} else {
err = *reinterpret_cast<const status_t*>(tr.data.ptr.buffer);
freeBuffer(nullptr,
reinterpret_cast<const uint8_t*>(tr.data.ptr.buffer),
tr.data_size,
reinterpret_cast<const binder_size_t*>(tr.data.ptr.offsets),
tr.offsets_size/sizeof(binder_size_t));
}
} else {
freeBuffer(nullptr,
reinterpret_cast<const uint8_t*>(tr.data.ptr.buffer),
tr.data_size,
reinterpret_cast<const binder_size_t*>(tr.data.ptr.offsets),
tr.offsets_size/sizeof(binder_size_t));
continue;
}
}
goto finish;
default:
err = executeCommand(cmd);
if (err != NO_ERROR) goto finish;
break;
}
}
finish:
if (err != NO_ERROR) {
if (acquireResult) *acquireResult = err;
if (reply) reply->setError(err);
mLastError = err;
}
return err;
}看下talkWithDriver：
代码语言：javascript复制status_t IPCThreadState::talkWithDriver(bool doReceive)
{
if (mProcess->mDriverFD < 0) {
return -EBADF;
}
binder_write_read bwr; // 被驱动读取的结构
// Is the read buffer empty?
const bool needRead = mIn.dataPosition() >= mIn.dataSize();
// We don't want to write anything if we are still reading
// from data left in the input buffer and the caller
// has requested to read the next data.
const size_t outAvail = (!doReceive || needRead) ? mOut.dataSize() : 0;
bwr.write_size = outAvail;
bwr.write_buffer = (uintptr_t)mOut.data();
// This is what we'll read.
if (doReceive && needRead) {
bwr.read_size = mIn.dataCapacity();
bwr.read_buffer = (uintptr_t)mIn.data();
} else {
bwr.read_size = 0;
bwr.read_buffer = 0;
}
...
// Return immediately if there is nothing to do.
if ((bwr.write_size == 0) && (bwr.read_size == 0)) return NO_ERROR;
bwr.write_consumed = 0;
bwr.read_consumed = 0;
status_t err;
...
#if defined(__ANDROID__)
if (ioctl(mProcess->mDriverFD, BINDER_WRITE_READ, &bwr) >= 0) //和驱动交互
err = NO_ERROR;
else
err = -errno;
#else
err = INVALID_OPERATION;
#endif
...
} while (err == -EINTR);
if (err >= NO_ERROR) {
if (bwr.write_consumed > 0) {
...
else {
mOut.setDataSize(0);
processPostWriteDerefs();
}
}
if (bwr.read_consumed > 0) {
mIn.setDataSize(bwr.read_consumed);
mIn.setDataPosition(0);
}
...
return NO_ERROR;
}
return err;
}这儿就是将请求写入binder_write_read，并和驱动进行交互。在进入驱动前，先看下binder_write_read的结构：
代码语言：javascript复制struct binder_write_read {
binder_size_t write_size; /* bytes to write */
binder_size_t write_consumed; /* bytes consumed by driver */
binder_uintptr_t write_buffer;
binder_size_t read_size; /* bytes to read */
binder_size_t read_consumed; /* bytes consumed by driver */
binder_uintptr_t read_buffer;
};在驱动中，ioctl的操作如下：
代码语言：javascript复制static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
int ret;
struct binder_proc *proc = filp->private_data;
struct binder_thread *thread;
unsigned int size = _IOC_SIZE(cmd);
void __user *ubuf = (void __user *)arg;
binder_selftest_alloc(&proc->alloc);
ret = wait_event_interruptible(binder_user_error_wait, binder_stop_on_user_error < 2);
if (ret)
goto err_unlocked;
thread = binder_get_thread(proc);
if (thread == NULL) {
ret = -ENOMEM;
goto err;
}
switch (cmd) {
case BINDER_WRITE_READ:
ret = binder_ioctl_write_read(filp, cmd, arg, thread);
if (ret)
goto err;
break;
...先看下binder_get_thread的操作：
代码语言：javascript复制static struct binder_thread *binder_get_thread(struct binder_proc *proc)
{
struct binder_thread *thread;
struct binder_thread *new_thread;
binder_inner_proc_lock(proc);
thread = binder_get_thread_ilocked(proc, NULL); // 从红黑树中查找binder线程，每个用户态线程都对应一个binder_thread
binder_inner_proc_unlock(proc);
if (!thread) {
new_thread = kzalloc(sizeof(*thread), GFP_KERNEL);
if (new_thread == NULL)
return NULL;
binder_inner_proc_lock(proc);
thread = binder_get_thread_ilocked(proc, new_thread);// 如果没找到，则将new_thread加入红黑树，并完成初始化
binder_inner_proc_unlock(proc);
if (thread != new_thread)
kfree(new_thread);
}
return thread;
}binder_get_thread_ilocked的逻辑如下：
代码语言：javascript复制static struct binder_thread *binder_get_thread_ilocked(
struct binder_proc *proc, struct binder_thread *new_thread)
{
struct binder_thread *thread = NULL;
struct rb_node *parent = NULL;
struct rb_node **p = &proc->threads.rb_node;
// 遍历红黑树，key是线程的pid，也可以理解成tid
while (*p) {
parent = *p;
thread = rb_entry(parent, struct binder_thread, rb_node);
if (current->pid < thread->pid)
p = &(*p)->rb_left;
else if (current->pid > thread->pid)
p = &(*p)->rb_right;
else
return thread;
}
if (!new_thread)
return NULL;
thread = new_thread; // 线程初始化
binder_stats_created(BINDER_STAT_THREAD);
thread->proc = proc;
thread->pid = current->pid;
atomic_set(&thread->tmp_ref, 0);
init_waitqueue_head(&thread->wait);
INIT_LIST_HEAD(&thread->todo);
rb_link_node(&thread->rb_node, parent, p); // 加入红黑树
rb_insert_color(&thread->rb_node, &proc->threads);
thread->looper_need_return = true;
thread->return_error.work.type = BINDER_WORK_RETURN_ERROR;
thread->return_error.cmd = BR_OK;
thread->reply_error.work.type = BINDER_WORK_RETURN_ERROR;
thread->reply_error.cmd = BR_OK;
INIT_LIST_HEAD(&new_thread->waiting_thread_node);
return thread;
}拿到线程后，接下来就是binder_ioctl_write_read：
代码语言：javascript复制static int binder_ioctl_write_read(struct file *filp,
unsigned int cmd, unsigned long arg,
struct binder_thread *thread)
{
int ret = 0;
struct binder_proc *proc = filp->private_data;
unsigned int size = _IOC_SIZE(cmd);
void __user *ubuf = (void __user *)arg;
struct binder_write_read bwr;
if (size != sizeof(struct binder_write_read)) {
ret = -EINVAL;
goto out;
}
// 拷贝用户态数据
if (copy_from_user(&bwr, ubuf, sizeof(bwr))) {
ret = -EFAULT;
goto out;
}
if (bwr.write_size > 0) {
ret = binder_thread_write(proc, thread, // 读取数据，包含了命令码等
bwr.write_buffer,
bwr.write_size,
&bwr.write_consumed);
trace_binder_write_done(ret);
if (ret < 0) {
bwr.read_consumed = 0;
if (copy_to_user(ubuf, &bwr, sizeof(bwr)))
ret = -EFAULT;
goto out;
}
}
...binder_thread_write里面就是解析命令内容了,只需要关注我们发的cmd是BC_TRANSACTION，code是PING_TRANSACTION：
代码语言：javascript复制static int binder_thread_write(struct binder_proc *proc,
struct binder_thread *thread,
binder_uintptr_t binder_buffer, size_t size,
binder_size_t *consumed)
{
uint32_t cmd;
struct binder_context *context = proc->context;
void __user *buffer = (void __user *)(uintptr_t)binder_buffer;
void __user *ptr = buffer *consumed;
void __user *end = buffer size;
...
case BC_TRANSACTION:
case BC_REPLY: {
struct binder_transaction_data tr;
if (copy_from_user(&tr, ptr, sizeof(tr)))
return -EFAULT;
ptr = sizeof(tr);
binder_transaction(proc, thread, &tr,
cmd == BC_REPLY, 0);
break;
}
...这儿就是读取内容，然后解析出命令，对于我们，命令是BC_TRANSACTION， 然后调用binder_transaction进行操作：
代码语言：javascript复制static void binder_transaction(struct binder_proc *proc,
struct binder_thread *thread,
struct binder_transaction_data *tr, int reply,
binder_size_t extra_buffers_size)
{
if (tr->target.handle) {
struct binder_ref *ref;
/*
* There must already be a strong ref
* on this node. If so, do a strong
* increment on the node to ensure it
* stays alive until the transaction is
* done.
*/
binder_proc_lock(proc);
ref = binder_get_ref_olocked(proc, tr->target.handle,
true);
if (ref) {
target_node = binder_get_node_refs_for_txn(
ref->node, &target_proc,
&return_error);
} else {
binder_user_error("%d:%d got transaction to invalid handlen",
proc->pid, thread->pid);
return_error = BR_FAILED_REPLY;
}
binder_proc_unlock(proc);
} else {
// 代码较长，略去本流程走不到的部分， handle是0，因此获取的是servicemanager
mutex_lock(&context->context_mgr_node_lock);
target_node = context->binder_context_mgr_node;
if (target_node)
target_node = binder_get_node_refs_for_txn(
target_node, &target_proc,
&return_error);
else
return_error = BR_DEAD_REPLY;
mutex_unlock(&context->context_mgr_node_lock);
if (target_node && target_proc->pid == proc->pid) {
binder_user_error("%d:%d got transaction to context manager from process owning itn",
proc->pid, thread->pid);
return_error = BR_FAILED_REPLY;
return_error_param = -EINVAL;
return_error_line = __LINE__;
goto err_invalid_target_handle;
}
}
...
//如果是同步操作，并且当前线程有binder事务，则在找目标线程的时候就可以尝试优化了。
//优化的场景主要是为了提升并法度，如果是B线程向A线程发一个请求，在A线程处理完之前，B线程应该是阻塞的，如果A线程这时候也需要让B线程所在的binder 处理一个请求，
//这时候直接让B线程来处理就行，不需要重新开一个线程了
if (!(tr->flags & TF_ONE_WAY) && thread->transaction_stack) {
struct binder_transaction *tmp;
tmp = thread->transaction_stack;
if (tmp->to_thread != thread) { // 线程的事物目标线程当然需要是自己
spin_lock(&tmp->lock);
binder_user_error("%d:%d got new transaction with bad transaction stack, transaction %d has target %d:%dn",
proc->pid, thread->pid, tmp->debug_id,
tmp->to_proc ? tmp->to_proc->pid : 0,
tmp->to_thread ?
tmp->to_thread->pid : 0);
spin_unlock(&tmp->lock);
binder_inner_proc_unlock(proc);
return_error = BR_FAILED_REPLY;
return_error_param = -EPROTO;
return_error_line = __LINE__;
goto err_bad_call_stack;
}
while (tmp) {
struct binder_thread *from; // 查找目标线程的优化，见上面注释
spin_lock(&tmp->lock);
from = tmp->from;
if (from && from->proc == target_proc) {
atomic_inc(&from->tmp_ref);
target_thread = from;
spin_unlock(&tmp->lock);
break;
}
spin_unlock(&tmp->lock);
tmp = tmp->from_parent;
}
}
...
} else if (!(t->flags & TF_ONE_WAY)) {
BUG_ON(t->buffer->async_transaction != 0);
binder_inner_proc_lock(proc);
/*
* Defer the TRANSACTION_COMPLETE, so we don't return to
* userspace immediately; this allows the target process to
* immediately start processing this transaction, reducing
* latency. We will then return the TRANSACTION_COMPLETE when
* the target replies (or there is an error).
*/
binder_enqueue_deferred_thread_work_ilocked(thread, tcomplete);
t->need_reply = 1;
t->from_parent = thread->transaction_stack;
thread->transaction_stack = t;
binder_inner_proc_unlock(proc);
if (!binder_proc_transaction(t, target_proc, target_thread)) { // 将binder事务添加到目标线程或进程的todo 队列中
binder_inner_proc_lock(proc);
binder_pop_transaction_ilocked(thread, t);
binder_inner_proc_unlock(proc);
goto err_dead_proc_or_thread;到这里，过去servicemanager的binder流程就结束了。最后再简单看下binder_proc_transaction的内容：
代码语言：javascript复制static bool binder_proc_transaction(struct binder_transaction *t,
struct binder_proc *proc,
struct binder_thread *thread)
{
struct binder_node *node = t->buffer->target_node;
bool oneway = !!(t->flags & TF_ONE_WAY);
bool pending_async = false;
BUG_ON(!node);
binder_node_lock(node);
if (oneway) {
BUG_ON(thread);
if (node->has_async_transaction)
pending_async = true;
else
node->has_async_transaction = true;
}
binder_inner_proc_lock(proc);
if (proc->is_dead || (thread && thread->is_dead)) {
binder_inner_proc_unlock(proc);
binder_node_unlock(node);
return false;
}
if (!thread && !pending_async)
thread = binder_select_thread_ilocked(proc); // 返回一个空闲线程
if (thread)
binder_enqueue_thread_work_ilocked(thread, &t->work); // 添加到空闲线程的队列中
else if (!pending_async)
binder_enqueue_work_ilocked(&t->work, &proc->todo); // 添加到目标进程的队列中
else
binder_enqueue_work_ilocked(&t->work, &node->async_todo); // 添加到异步队列中
if (!pending_async)
binder_wakeup_thread_ilocked(proc, thread, !oneway /* sync */); // 唤醒目标线程
binder_inner_proc_unlock(proc);
binder_node_unlock(node);
return true;
}介绍到这里，defaultServiceManager涉及的流程就介绍完了，最后用一个流程图总结下：
binder.png
本篇总结
本篇介绍了下servicemanager proxy的获取流程，涉及了ProcessState(进程单例)， IPCThreadState(线程单例)。 binder驱动的open，mmap，ioctl部分。该流程比起其他调用流程稍微简单一些，不过对于熟悉binder 工作流程还是很有帮助的。
