Android Framework学习(十三)之启动ServiceManager

ServiceManager是Binder IPC通信过程中的守护进程，本身也是一个Binder服务，但并没有采用libbinder中的多线程模型来与Binder驱动通信，而是自行编写了binder.c直接和Binder驱动来通信，并且只有一个循环binder_loop来进行读取和处理事务，这样的好处是简单而高效。

ServiceManager本身工作相对简单，其功能：查询和注册服务。 对于Binder IPC通信过程中，其实更多的情形是BpBinder和BBinder之间的通信，比如ActivityManagerProxy和ActivityManagerService之间的通信等。

启动过程主要以下几个阶段：

1.打开binder驱动：binder_open；
2.注册成为binder服务的大管家：binder_become_context_manager；
3.进入无限循环，处理client端发来的请求：binder_loop；

启动过程

ServiceManager是由init进程通过解析init.rc文件而创建的，其所对应的可执行程序/system/bin/servicemanager，所对应的源文件是service_manager.c，进程名为/system/bin/servicemanager。

service servicemanager /system/bin/servicemanager
    class core
    user system
    group system
    critical
    onrestart restart healthd
    onrestart restart zygote
    onrestart restart media
    onrestart restart surfaceflinger
    onrestart restart drm

启动Service Manager的入口函数是service_manager.c中的main()方法，代码如下：

main
service_manager.c

int main(int argc, char **argv)
{
    struct binder_state *bs;
    //打开binder驱动，申请128k字节大小的内存空间 
    bs = binder_open(128*1024);
    ...

    //成为上下文管理者 
    if (binder_become_context_manager(bs)) {
        return -1;
    }

    selinux_enabled = is_selinux_enabled(); //selinux权限是否使能
    sehandle = selinux_android_service_context_handle();
    selinux_status_open(true);

    if (selinux_enabled > 0) {
        if (sehandle == NULL) {  
            abort(); //无法获取sehandle
        }
        if (getcon(&service_manager_context) != 0) {
            abort(); //无法获取service_manager上下文
        }
    }
    ...

    //进入无限循环，处理client端发来的请求 
    binder_loop(bs, svcmgr_handler);
    return 0;
}

binder_open
servicemanager/binder.c

struct binder_state *binder_open(size_t mapsize)
{
    struct binder_state *bs;
    struct binder_version vers;

    bs = malloc(sizeof(*bs));
    if (!bs) {
        errno = ENOMEM;
        return NULL;
    }

    //通过系统调用陷入内核，打开Binder设备驱动
    bs->fd = open("/dev/binder", O_RDWR);
    if (bs->fd < 0) {
        goto fail_open; // 无法打开binder设备
    }

     //通过系统调用，ioctl获取binder版本信息
    if ((ioctl(bs->fd, BINDER_VERSION, &vers) == -1) ||
        (vers.protocol_version != BINDER_CURRENT_PROTOCOL_VERSION)) {
        goto fail_open; //内核空间与用户空间的binder不是同一版本
    }

    bs->mapsize = mapsize;
    //通过系统调用，mmap内存映射，mmap必须是page的整数倍
    bs->mapped = mmap(NULL, mapsize, PROT_READ, MAP_PRIVATE, bs->fd, 0);
    if (bs->mapped == MAP_FAILED) {
        goto fail_map; // binder设备内存无法映射
    }

    return bs;

fail_map:
    close(bs->fd);
fail_open:
    free(bs);
    return NULL;
}

打开binder驱动相关操作:

先调用open()打开binder设备，open()方法经过系统调用，进入Binder驱动，然后调用方法binder_open()，该方法会在Binder驱动层创建一个binder_proc对象，再将binder_proc对象赋值给fd->private_data，同时放入全局链表binder_procs。再通过ioctl()检验当前binder版本与Binder驱动层的版本是否一致。

调用mmap()进行内存映射，同理mmap()方法经过系统调用，对应于Binder驱动层的binder_mmap()方法，该方法会在Binder驱动层创建Binder_buffer对象，并放入当前binder_proc的proc->buffers链表。

binder_state
servicemanager/binder.c

struct binder_state
{
    int fd; // dev/binder的文件描述符
    void *mapped; //指向mmap的内存地址
    size_t mapsize; //分配的内存大小，默认为128KB
};

binder_become_context_manager
servicemanager/binder.c

int binder_become_context_manager(struct binder_state *bs)
{
    //通过ioctl，传递BINDER_SET_CONTEXT_MGR指令
    return ioctl(bs->fd, BINDER_SET_CONTEXT_MGR, 0);
}

成为上下文的管理者，整个系统中只有一个这样的管理者。 通过ioctl()方法经过系统调用，对应于Binder驱动层的binder_ioctl()方法.

binder_ioctl
kernel/drivers/android/binder.c

static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
    binder_lock(__func__);
    switch (cmd) {
      case BINDER_SET_CONTEXT_MGR:
          ret = binder_ioctl_set_ctx_mgr(filp);
          break;
      }
      case :...
    }
    binder_unlock(__func__);
}

根据参数BINDER_SET_CONTEXT_MGR，最终调用binder_ioctl_set_ctx_mgr()方法，这个过程会持有binder_main_lock。

binder_ioctl_set_ctx_mgr
kernel/drivers/android/binder.c

static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
    binder_lock(__func__);
    switch (cmd) {
      case BINDER_SET_CONTEXT_MGR:
          ret = binder_ioctl_set_ctx_mgr(filp);
          break;
      }
      case :...
    }
    binder_unlock(__func__);
}

根据参数BINDER_SET_CONTEXT_MGR，最终调用binder_ioctl_set_ctx_mgr()方法，这个过程会持有binder_main_lock。

binder_ioctl_set_ctx_mgr
kernel/drivers/android/binder.c

static int binder_ioctl_set_ctx_mgr(struct file *filp)
{
    int ret = 0;
    struct binder_proc *proc = filp->private_data;
    kuid_t curr_euid = current_euid();

    //保证只创建一次mgr_node对象
    if (binder_context_mgr_node != NULL) {
        ret = -EBUSY; 
        goto out;
    }

    if (uid_valid(binder_context_mgr_uid)) {
        ...
    } else {
        //设置当前线程euid作为Service Manager的uid
        binder_context_mgr_uid = curr_euid;
    }

    //创建ServiceManager实体
    binder_context_mgr_node = binder_new_node(proc, 0, 0);
    ...
    binder_context_mgr_node->local_weak_refs++;
    binder_context_mgr_node->local_strong_refs++;
    binder_context_mgr_node->has_strong_ref = 1;
    binder_context_mgr_node->has_weak_ref = 1;
out:
    return ret;
}

进入binder驱动，在Binder驱动中定义的静态变量

// service manager所对应的binder_node;
static struct binder_node *binder_context_mgr_node;
// 运行service manager的线程uid
static kuid_t binder_context_mgr_uid = INVALID_UID;

创建了全局的binder_node对象binder_context_mgr_node，并将binder_context_mgr_node的强弱引用各加1.

binder_new_node
kernel/drivers/android/binder.c

static struct binder_node *binder_new_node(struct binder_proc *proc,
                       binder_uintptr_t ptr,
                       binder_uintptr_t cookie)
{
    struct rb_node **p = &proc->nodes.rb_node;
    struct rb_node *parent = NULL;
    struct binder_node *node;
    //首次进来为空
    while (*p) {
        parent = *p;
        node = rb_entry(parent, struct binder_node, rb_node);

        if (ptr < node->ptr)
            p = &(*p)->rb_left;
        else if (ptr > node->ptr)
            p = &(*p)->rb_right;
        else
            return NULL;
    }

    //给新创建的binder_node 分配内核空间
    node = kzalloc(sizeof(*node), GFP_KERNEL);
    if (node == NULL)
        return NULL;
    binder_stats_created(BINDER_STAT_NODE);
    // 将新创建的node对象添加到proc红黑树；
    rb_link_node(&node->rb_node, parent, p);
    rb_insert_color(&node->rb_node, &proc->nodes);
    node->debug_id = ++binder_last_id;
    node->proc = proc;
    node->ptr = ptr;
    node->cookie = cookie;
    node->work.type = BINDER_WORK_NODE; //设置binder_work的type
    INIT_LIST_HEAD(&node->work.entry);
    INIT_LIST_HEAD(&node->async_todo);
    return node;
}

在Binder驱动层创建binder_node结构体对象，并将当前binder_proc加入到binder_node的node->proc。并创建binder_node的async_todo和binder_work两个队列。
binder_loop
servicemanager/binder.c

void binder_loop(struct binder_state *bs, binder_handler func)
{
    int res;
    struct binder_write_read bwr;
    uint32_t readbuf[32];

    bwr.write_size = 0;
    bwr.write_consumed = 0;
    bwr.write_buffer = 0;

    readbuf[0] = BC_ENTER_LOOPER;
    //将BC_ENTER_LOOPER命令发送给binder驱动，让Service Manager进入循环 
    binder_write(bs, readbuf, sizeof(uint32_t));

    for (;;) {
        bwr.read_size = sizeof(readbuf);
        bwr.read_consumed = 0;
        bwr.read_buffer = (uintptr_t) readbuf;

        res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr); //进入循环，不断地binder读写过程
        if (res < 0) {
            break;
        }

        // 解析binder信息 
        res = binder_parse(bs, 0, (uintptr_t) readbuf, bwr.read_consumed, func);
        if (res == 0) {
            break;
        }
        if (res < 0) {
            break;
        }
    }
}

进入循环读写操作，由main()方法传递过来的参数func指向svcmgr_handler。

binder_write通过ioctl()将BC_ENTER_LOOPER命令发送给binder驱动，此时bwr只有write_buffer有数据，进入binder_thread_write()方法。 接下来进入for循环，执行ioctl()，此时bwr只有read_buffer有数据，那么进入binder_thread_read()方法。

binder_write
servicemanager/binder.c

int binder_write(struct binder_state *bs, void *data, size_t len)
{
    struct binder_write_read bwr;
    int res;

    bwr.write_size = len;
    bwr.write_consumed = 0;
    bwr.write_buffer = (uintptr_t) data; //此处data为BC_ENTER_LOOPER
    bwr.read_size = 0;
    bwr.read_consumed = 0;
    bwr.read_buffer = 0;
    res = ioctl(bs->fd, BINDER_WRITE_READ, &bwr);
    return res;
}

根据传递进来的参数，初始化bwr，其中write_size大小为4，write_buffer指向缓冲区的起始地址，其内容为BC_ENTER_LOOPER请求协议号。通过ioctl将bwr数据发送给binder驱动，则调用其binder_ioctl方法，如下：
binder_ioctl
kernel/drivers/android/binder.c

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;
    ret = wait_event_interruptible(binder_user_error_wait, binder_stop_on_user_error < 2);
    ...

    binder_lock(__func__);
    thread = binder_get_thread(proc); //获取binder_thread
    switch (cmd) {
      case BINDER_WRITE_READ:  //进行binder的读写操作
          ret = binder_ioctl_write_read(filp, cmd, arg, thread); 
          if (ret)
              goto err;
          break;
      case ...
    }
    ret = 0;

err:
    if (thread)
        thread->looper &= ~BINDER_LOOPER_STATE_NEED_RETURN;
    binder_unlock(__func__);
     ...
    return ret;
}

binder_ioctl_write_read
kernel/drivers/android/binder.c

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;
    void __user *ubuf = (void __user *)arg;
    struct binder_write_read bwr;

    if (copy_from_user(&bwr, ubuf, sizeof(bwr))) { //把用户空间数据ubuf拷贝到bwr
        ret = -EFAULT;
        goto out;
    }

    if (bwr.write_size > 0) { //此时写缓存有数据
        ret = binder_thread_write(proc, thread,
                  bwr.write_buffer, bwr.write_size, &bwr.write_consumed);
        ...
    }

    if (bwr.read_size > 0) { //此时读缓存有数据
        ...
    }

    if (copy_to_user(ubuf, &bwr, sizeof(bwr))) { //将内核数据bwr拷贝到用户空间ubuf
        ret = -EFAULT;
        goto out;
    }
out:
    return ret;
}

此处将用户空间的binder_write_read结构体 拷贝到内核空间.
binder_thread_write
kernel/drivers/android/binder.c

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;
  void __user *buffer = (void __user *)(uintptr_t)binder_buffer;
  void __user *ptr = buffer + *consumed;
  void __user *end = buffer + size;

  while (ptr < end && thread->return_error == BR_OK) {
    get_user(cmd, (uint32_t __user *)ptr); //获取命令
    switch (cmd) {
      case BC_ENTER_LOOPER:
          //设置该线程的looper状态
          thread->looper |= BINDER_LOOPER_STATE_ENTERED;
          break;
      case ...;
    }
  }    
 }

binder_parse
servicemanager/binder.c

int binder_parse(struct binder_state *bs, struct binder_io *bio,
                 uintptr_t ptr, size_t size, binder_handler func)
{
    int r = 1;
    uintptr_t end = ptr + (uintptr_t) size;

    while (ptr < end) {
        uint32_t cmd = *(uint32_t *) ptr;
        ptr += sizeof(uint32_t);
        switch(cmd) {
        case BR_NOOP:  //无操作，退出循环
            break;
        case BR_TRANSACTION_COMPLETE:
            break;
        case BR_INCREFS:
        case BR_ACQUIRE:
        case BR_RELEASE:
        case BR_DECREFS:
            ptr += sizeof(struct binder_ptr_cookie);
            break;
        case BR_TRANSACTION: {
            struct binder_transaction_data *txn = (struct binder_transaction_data *) ptr;
            ...
            binder_dump_txn(txn);
            if (func) {
                unsigned rdata[256/4];
                struct binder_io msg; 
                struct binder_io reply;
                int res;

                bio_init(&reply, rdata, sizeof(rdata), 4);
                bio_init_from_txn(&msg, txn); //从txn解析出binder_io信息

                res = func(bs, txn, &msg, &reply);

                binder_send_reply(bs, &reply, txn->data.ptr.buffer, res);
            }
            ptr += sizeof(*txn);
            break;
        }
        case BR_REPLY: {
            struct binder_transaction_data *txn = (struct binder_transaction_data *) ptr;
            ...
            binder_dump_txn(txn);
            if (bio) {
                bio_init_from_txn(bio, txn);
                bio = 0;
            }
            ptr += sizeof(*txn);
            r = 0;
            break;
        }
        case BR_DEAD_BINDER: {
            struct binder_death *death = (struct binder_death *)(uintptr_t) *(binder_uintptr_t *)ptr;
            ptr += sizeof(binder_uintptr_t);
            // binder死亡消息
            death->func(bs, death->ptr);
            break;
        }
        case BR_FAILED_REPLY:
            r = -1;
            break;
        case BR_DEAD_REPLY:
            r = -1;
            break;
        default:
            return -1;
        }
    }
    return r;
}

解析binder信息，此处参数ptr指向BC_ENTER_LOOPER，func指向svcmgr_handler。故有请求到来，则调用svcmgr_handler。

bio_init
servicemanager/binder.c

void bio_init(struct binder_io *bio, void *data,
              size_t maxdata, size_t maxoffs)
{
    size_t n = maxoffs * sizeof(size_t);
    if (n > maxdata) {
        ...
    }

    bio->data = bio->data0 = (char *) data + n;
    bio->offs = bio->offs0 = data;
    bio->data_avail = maxdata - n;
    bio->offs_avail = maxoffs;
    bio->flags = 0;
}

其中

struct binder_io
{
    char *data;            /* pointer to read/write from */
    binder_size_t *offs;   /* array of offsets */
    size_t data_avail;     /* bytes available in data buffer */
    size_t offs_avail;     /* entries available in offsets array */

    char *data0;           //data buffer起点位置
    binder_size_t *offs0;  //buffer偏移量的起点位置
    uint32_t flags;
    uint32_t unused;
};

bio_init_from_txn
servicemanager/binder.c

void bio_init_from_txn(struct binder_io *bio, struct binder_transaction_data *txn)
{
    bio->data = bio->data0 = (char *)(intptr_t)txn->data.ptr.buffer;
    bio->offs = bio->offs0 = (binder_size_t *)(intptr_t)txn->data.ptr.offsets;
    bio->data_avail = txn->data_size;
    bio->offs_avail = txn->offsets_size / sizeof(size_t);
    bio->flags = BIO_F_SHARED;
}

将readbuf的数据赋给bio对象的data
svcmgr_handler
service_manager.c

int svcmgr_handler(struct binder_state *bs,
                   struct binder_transaction_data *txn,
                   struct binder_io *msg,
                   struct binder_io *reply)
{
    struct svcinfo *si; //
    uint16_t *s;
    size_t len;
    uint32_t handle;
    uint32_t strict_policy;
    int allow_isolated;
    ...

    strict_policy = bio_get_uint32(msg);
    s = bio_get_string16(msg, &len);
    ...

    switch(txn->code) {
    case SVC_MGR_GET_SERVICE:
    case SVC_MGR_CHECK_SERVICE: 
        s = bio_get_string16(msg, &len); //服务名
        //根据名称查找相应服务 
        handle = do_find_service(bs, s, len, txn->sender_euid, txn->sender_pid);
        //
        bio_put_ref(reply, handle);
        return 0;

    case SVC_MGR_ADD_SERVICE: 
        s = bio_get_string16(msg, &len); //服务名
        handle = bio_get_ref(msg); //handle
        allow_isolated = bio_get_uint32(msg) ? 1 : 0;
         //注册指定服务 
        if (do_add_service(bs, s, len, handle, txn->sender_euid,
            allow_isolated, txn->sender_pid))
            return -1;
        break;

    case SVC_MGR_LIST_SERVICES: {  
        uint32_t n = bio_get_uint32(msg);

        if (!svc_can_list(txn->sender_pid)) {
            return -1;
        }
        si = svclist;
        while ((n-- > 0) && si)
            si = si->next;
        if (si) {
            bio_put_string16(reply, si->name);
            return 0;
        }
        return -1;
    }
    default:
        return -1;
    }

    bio_put_uint32(reply, 0);
    return 0;
}

该方法的功能：查询服务，注册服务，以及列举所有服务

svcinfo

struct svcinfo
{
    struct svcinfo *next;
    uint32_t handle; //服务的handle值
    struct binder_death death;
    int allow_isolated;
    size_t len; //名字长度
    uint16_t name[0]; //服务名
};

每一个服务用svcinfo结构体来表示，该handle值是在注册服务的过程中，由服务所在进程那一端所确定的。

核心工作

servicemanager的核心工作就是注册服务和查询服务。
do_find_service
service_manager.c

uint32_t do_find_service(struct binder_state *bs, const uint16_t *s, size_t len, uid_t uid, pid_t spid)
{
    //查询相应的服务 
    struct svcinfo *si = find_svc(s, len);

    if (!si || !si->handle) {
        return 0;
    }

    if (!si->allow_isolated) {
        uid_t appid = uid % AID_USER;
        //检查该服务是否允许孤立于进程而单独存在
        if (appid >= AID_ISOLATED_START && appid <= AID_ISOLATED_END) {
            return 0;
        }
    }

    //服务是否满足查询条件
    if (!svc_can_find(s, len, spid)) {
        return 0;
    }
    return si->handle;
}

查询到目标服务，并返回该服务所对应的handle

find_svc

struct svcinfo *find_svc(const uint16_t *s16, size_t len)
{
    struct svcinfo *si;

    for (si = svclist; si; si = si->next) {
        //当名字完全一致，则返回查询到的结果
        if ((len == si->len) &&
            !memcmp(s16, si->name, len * sizeof(uint16_t))) {
            return si;
        }
    }
    return NULL;
}

从svclist服务列表中，根据服务名遍历查找是否已经注册。当服务已存在svclist，则返回相应的服务名，否则返回NULL。

当找到服务的handle, 则调用bio_put_ref(reply, handle)，将handle封装到reply.
bio_put_ref

void bio_put_ref(struct binder_io *bio, uint32_t handle)
{
    struct flat_binder_object *obj;

    if (handle)
        obj = bio_alloc_obj(bio); 
    else
        obj = bio_alloc(bio, sizeof(*obj));

    if (!obj)
        return;

    obj->flags = 0x7f | FLAT_BINDER_FLAG_ACCEPTS_FDS;
    obj->type = BINDER_TYPE_HANDLE; //返回的是HANDLE类型
    obj->handle = handle;
    obj->cookie = 0;
}

bio_alloc_obj

static struct flat_binder_object *bio_alloc_obj(struct binder_io *bio)
{
    struct flat_binder_object *obj;
    obj = bio_alloc(bio, sizeof(*obj));

    if (obj && bio->offs_avail) {
        bio->offs_avail--;
        *bio->offs++ = ((char*) obj) - ((char*) bio->data0);
        return obj;
    }
    bio->flags |= BIO_F_OVERFLOW;
    return NULL;
}

bio_alloc

static void *bio_alloc(struct binder_io *bio, size_t size)
{
    size = (size + 3) & (~3);
    if (size > bio->data_avail) {
        bio->flags |= BIO_F_OVERFLOW;
        return NULL;
    } else {
        void *ptr = bio->data;
        bio->data += size;
        bio->data_avail -= size;
        return ptr;
    }
}

do_add_service
service_manager.c

int do_add_service(struct binder_state *bs,
                   const uint16_t *s, size_t len,
                   uint32_t handle, uid_t uid, int allow_isolated,
                   pid_t spid)
{
    struct svcinfo *si;

    if (!handle || (len == 0) || (len > 127))
        return -1;

    //权限检查
    if (!svc_can_register(s, len, spid)) {
        return -1;
    }

    //服务检索
    si = find_svc(s, len);
    if (si) {
        if (si->handle) {
            svcinfo_death(bs, si); //服务已注册时，释放相应的服务
        }
        si->handle = handle;
    } else {
        si = malloc(sizeof(*si) + (len + 1) * sizeof(uint16_t));
        if (!si) {  //内存不足，无法分配足够内存
            return -1;
        }
        si->handle = handle;
        si->len = len;
        memcpy(si->name, s, (len + 1) * sizeof(uint16_t)); //内存拷贝服务信息
        si->name[len] = '\0';
        si->death.func = (void*) svcinfo_death;
        si->death.ptr = si;
        si->allow_isolated = allow_isolated;
        si->next = svclist; // svclist保存所有已注册的服务
        svclist = si;
    }

    //以BC_ACQUIRE命令，handle为目标的信息，通过ioctl发送给binder驱动
    binder_acquire(bs, handle);
    //以BC_REQUEST_DEATH_NOTIFICATION命令的信息，通过ioctl发送给binder驱动，主要用于清理内存等收尾工作。
    binder_link_to_death(bs, handle, &si->death);
    return 0;
}

注册服务的分以下3部分工作：

svc_can_register：检查权限，检查selinux权限是否满足；
find_svc：服务检索，根据服务名来查询匹配的服务；
svcinfo_death：释放服务，当查询到已存在同名的服务，则先清理该服务信息，再将当前的服务加入到服务列表svclist；

svc_can_register
service_manager.c

static int svc_can_register(const uint16_t *name, size_t name_len, pid_t spid)
{
    const char *perm = "add";
    //检查selinux权限是否满足
    return check_mac_perms_from_lookup(spid, perm, str8(name, name_len)) ? 1 : 0;
}

svcinfo_death
service_manager.c

void svcinfo_death(struct binder_state *bs, void *ptr)
{
    struct svcinfo *si = (struct svcinfo* ) ptr;

    if (si->handle) {
        binder_release(bs, si->handle);
        si->handle = 0;
    }
}

bio_get_ref
servicemanager/binder.c

uint32_t bio_get_ref(struct binder_io *bio)
{
    struct flat_binder_object *obj;

    obj = _bio_get_obj(bio);
    if (!obj)
        return 0;

    if (obj->type == BINDER_TYPE_HANDLE)
        return obj->handle;

    return 0;
}

binder_link_to_death
servicemanager/binder.c

void binder_link_to_death(struct binder_state *bs, uint32_t target, struct binder_death *death)
{
    struct {
        uint32_t cmd;
        struct binder_handle_cookie payload;
    } __attribute__((packed)) data;

    data.cmd = BC_REQUEST_DEATH_NOTIFICATION;
    data.payload.handle = target;
    data.payload.cookie = (uintptr_t) death;
    binder_write(bs, &data, sizeof(data)); 
}

binder_write进入Binder driver后,直接调用后进入binder_thread_write, 处理BC_REQUEST_DEATH_NOTIFICATION命令
binder_ioctl_write_read
kernel/drivers/android/binder.c

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;
    void __user *ubuf = (void __user *)arg;
    struct binder_write_read bwr;

    if (copy_from_user(&bwr, ubuf, sizeof(bwr))) { //把用户空间数据ubuf拷贝到bwr
        ret = -EFAULT;
        goto out;
    }
    if (bwr.write_size > 0) { //此时写缓存有数据
        ret = binder_thread_write(proc, thread,
                  bwr.write_buffer, bwr.write_size, &bwr.write_consumed);
         if (ret < 0) {
              bwr.read_consumed = 0;
              if (copy_to_user(ubuf, &bwr, sizeof(bwr)))
                  ret = -EFAULT;
              goto out;
          }
    }

    if (bwr.read_size > 0) { //此时读缓存有数据
        ret = binder_thread_read(proc, thread, bwr.read_buffer,
                 bwr.read_size,
                 &bwr.read_consumed,
                 filp->f_flags & O_NONBLOCK);
        if (!list_empty(&proc->todo))  //进程todo队列不为空,则唤醒该队列中的线程
            wake_up_interruptible(&proc->wait);
        if (ret < 0) {
            if (copy_to_user(ubuf, &bwr, sizeof(bwr)))
                ret = -EFAULT;
            goto out;
        }
    }

    if (copy_to_user(ubuf, &bwr, sizeof(bwr))) { //将内核数据bwr拷贝到用户空间ubuf
        ret = -EFAULT;
        goto out;
    }
out:
    return ret;
}

binder_thread_write
kernel/drivers/android/binder.c

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; //ptr指向bwr中write_buffer的data.
  void __user *end = buffer + size;
  while (ptr < end && thread->return_error == BR_OK) {
    get_user(cmd, (uint32_t __user *)ptr); //获取BC_REQUEST_DEATH_NOTIFICATION
    ptr += sizeof(uint32_t);
    switch (cmd) {
        case BC_REQUEST_DEATH_NOTIFICATION:{ //注册死亡通知
            uint32_t target;
            void __user *cookie;
            struct binder_ref *ref;
            struct binder_ref_death *death;

            get_user(target, (uint32_t __user *)ptr); //获取target
            ptr += sizeof(uint32_t);
            get_user(cookie, (void __user * __user *)ptr); //获取death
            ptr += sizeof(void *);

            ref = binder_get_ref(proc, target); //拿到目标服务的binder_ref

            if (cmd == BC_REQUEST_DEATH_NOTIFICATION) {
                if (ref->death) {
                    break;  //已设置死亡通知
                }
                death = kzalloc(sizeof(*death), GFP_KERNEL);

                INIT_LIST_HEAD(&death->work.entry);
                death->cookie = cookie;
                ref->death = death;
                if (ref->node->proc == NULL) { //当目标binder服务所在进程已死,则发送死亡通知
                    ref->death->work.type = BINDER_WORK_DEAD_BINDER;
                    //当前线程为binder线程,则直接添加到当前线程的todo队列
                    if (thread->looper & (BINDER_LOOPER_STATE_REGISTERED | BINDER_LOOPER_STATE_ENTERED)) {
                        list_add_tail(&ref->death->work.entry, &thread->todo);
                    } else {
                        list_add_tail(&ref->death->work.entry, &proc->todo);
                        wake_up_interruptible(&proc->wait);
                    }
                }
            } else {
                ...
            }
        } break;
      case ...;
    }
    *consumed = ptr - buffer;
  }    }

此方法中的proc, thread都是指当前servicemanager进程的信息. 此时TODO队列有数据,则进入binder_thread_read.

那么哪些场景会向队列增加BINDER_WORK_DEAD_BINDER事务呢? 那就是当binder所在进程死亡后,会调用binder_release方法, 然后调用binder_node_release.这个过程便会发出死亡通知的回调.

binder_thread_read

static int binder_thread_read(struct binder_proc *proc,
                  struct binder_thread *thread,
                  binder_uintptr_t binder_buffer, size_t size,
                  binder_size_t *consumed, int non_block)
    ...
    //只有当前线程todo队列为空，并且transaction_stack也为空，才会开始处于当前进程的事务
    if (wait_for_proc_work) {
        ...
        ret = wait_event_freezable_exclusive(proc->wait, binder_has_proc_work(proc, thread));
    } else {
        ...
        ret = wait_event_freezable(thread->wait, binder_has_thread_work(thread));
    }
    binder_lock(__func__); //加锁

    if (wait_for_proc_work)
        proc->ready_threads--; //空闲的binder线程减1
    thread->looper &= ~BINDER_LOOPER_STATE_WAITING;

    while (1) {
        uint32_t cmd;
        struct binder_transaction_data tr;
        struct binder_work *w;
        struct binder_transaction *t = NULL;

        //从todo队列拿出前面放入的binder_work, 此时type为BINDER_WORK_DEAD_BINDER
        if (!list_empty(&thread->todo)) {
            w = list_first_entry(&thread->todo, struct binder_work,
                         entry);
        } else if (!list_empty(&proc->todo) && wait_for_proc_work) {
            w = list_first_entry(&proc->todo, struct binder_work,
                         entry);
        }

        switch (w->type) {
            case BINDER_WORK_DEAD_BINDER: {
              struct binder_ref_death *death;
              uint32_t cmd;

              death = container_of(w, struct binder_ref_death, work);
              if (w->type == BINDER_WORK_CLEAR_DEATH_NOTIFICATION)
                  ...
              else
                  cmd = BR_DEAD_BINDER; //进入此分支
              put_user(cmd, (uint32_t __user *)ptr);//拷贝到用户空间
              ptr += sizeof(uint32_t);

              //此处的cookie是前面传递的svcinfo_death
              put_user(death->cookie, (binder_uintptr_t __user *)ptr);
              ptr += sizeof(binder_uintptr_t);

              if (w->type == BINDER_WORK_CLEAR_DEATH_NOTIFICATION) {
                  ...
              } else
                  list_move(&w->entry, &proc->delivered_death);
              if (cmd == BR_DEAD_BINDER)
                  goto done;
            } break;
        }
    }
    ...
    return 0;
}

将命令BR_DEAD_BINDER写到用户空间, 此处的cookie是前面传递的svcinfo_death. 当binder_loop下一次 执行binder_parse的过程便会处理该消息。

binder_parse
servicemanager/binder.c

int binder_parse(struct binder_state *bs, struct binder_io *bio,
                 uintptr_t ptr, size_t size, binder_handler func)
{
    int r = 1;
    uintptr_t end = ptr + (uintptr_t) size;

    while (ptr < end) {
        uint32_t cmd = *(uint32_t *) ptr;
        ptr += sizeof(uint32_t);
        switch(cmd) {
            case BR_DEAD_BINDER: {
                struct binder_death *death = (struct binder_death *)(uintptr_t) *(binder_uintptr_t *)ptr;
                ptr += sizeof(binder_uintptr_t);
                // binder死亡消息
                death->func(bs, death->ptr);
                break;
            }
            ...
        }
    }
    return r;
}

death->func便是执行svcinfo_death()方法.
svcinfo_death
service_manager.c

void svcinfo_death(struct binder_state *bs, void *ptr)
{
    struct svcinfo *si = (struct svcinfo* ) ptr;

    if (si->handle) {
        binder_release(bs, si->handle);
        si->handle = 0;
    }
}

binder_release
service_manager.c

void binder_release(struct binder_state *bs, uint32_t target)
{
    uint32_t cmd[2];
    cmd[0] = BC_RELEASE;
    cmd[1] = target;
    binder_write(bs, cmd, sizeof(cmd));
}

向Binder Driver写入BC_RELEASE命令, 最终进入Binder Driver后执行binder_dec_ref(ref, 1)来减少binder node的引用.

binder_send_reply
servicemanager/binder.c

void binder_send_reply(struct binder_state *bs,
                       struct binder_io *reply,
                       binder_uintptr_t buffer_to_free,
                       int status)
{
    struct {
        uint32_t cmd_free;
        binder_uintptr_t buffer;
        uint32_t cmd_reply;
        struct binder_transaction_data txn;
    } __attribute__((packed)) data;

    data.cmd_free = BC_FREE_BUFFER; //free buffer命令
    data.buffer = buffer_to_free;
    data.cmd_reply = BC_REPLY; // reply命令
    data.txn.target.ptr = 0;
    data.txn.cookie = 0;
    data.txn.code = 0;
    if (status) {
        data.txn.flags = TF_STATUS_CODE;
        data.txn.data_size = sizeof(int);
        data.txn.offsets_size = 0;
        data.txn.data.ptr.buffer = (uintptr_t)&status;
        data.txn.data.ptr.offsets = 0;
    } else {
        data.txn.flags = 0;
        data.txn.data_size = reply->data - reply->data0;
        data.txn.offsets_size = ((char*) reply->offs) - ((char*) reply->offs0);
        data.txn.data.ptr.buffer = (uintptr_t)reply->data0;
        data.txn.data.ptr.offsets = (uintptr_t)reply->offs0;
    }
    //向Binder驱动通信
    binder_write(bs, &data, sizeof(data));
}

当执行binder_parse方法，先调用svcmgr_handler()，再然后执行binder_send_reply过程。该方法会调用 binder_write进入binder驱动后，将BC_FREE_BUFFER和BC_REPLY命令协议发送给Binder驱动，向client端发送reply. 其中data的数据区中保存的是TYPE为HANDLE.

总结

ServiceManger集中管理系统内的所有服务，通过权限控制进程是否有权注册服务,通过字符串名称来查找对应的Service; 由于ServiceManger进程建立跟所有向其注册服务的死亡通知, 那么当服务所在进程死亡后, 会只需告知ServiceManager. 每个Client通过查询ServiceManager可获取Server进程的情况，降低所有Client进程直接检测会导致负载过重。

ServiceManager启动流程：
打开binder驱动，并调用mmap()方法分配128k的内存映射空间：binder_open();
通知binder驱动使其成为守护进程：binder_become_context_manager()；
验证selinux权限，判断进程是否有权注册或查看指定服务；
进入循环状态，等待Client端的请求：binder_loop()。
注册服务的过程，根据服务名称，但同一个服务已注册，重新注册前会先移除之前的注册信息；
死亡通知: 当binder所在进程死亡后,会调用binder_release方法,然后调用binder_node_release.这个过程便会发出死亡通知的回调.
ServiceManager最核心的两个功能为查询和注册服务：

注册服务：记录服务名和handle信息，保存到svclist列表；
查询服务：根据服务名查询相应的的handle信息。