Binder机制情景分析之transaction_stack

一. 概述

这里以注册服务为例,当led_control_service请求注册服务时是通过handle找到的ServiceManager,但是ServiceManager是如何找到led_control_service进行回复的呢?

答:这里驱动中用到了一个传送栈记录了发送和接收进程,线程的信息; 接下来我们讲下具体的流程;

二. transaction_stack的来源

主要代码在binder_transaction()和binder_thread_read()中;

2.1 BC_TRANSACTION

从led_control_service调用BC_TRANSACTION开始,此时驱动会调用binder_transaction()函数;

如下是该函数中的一段代码,这段代码前面在深入驱动时未讲解:

    if (!reply && !(tr->flags & TF_ONE_WAY))                  ①
        t->from = thread;
    else
        t->from = NULL;
    t->sender_euid = task_euid(proc->tsk);
    t->to_proc = target_proc;                                 ②
    t->to_thread = target_thread;
    t->code = tr->code;
    t->flags = tr->flags;
    t->priority = task_nice(current);

①: 判断是否需要回复,需要则记录下当前进程的thread信息;
②: 记录下要目标进程信息和线程信息;

这里t是struct binder_transaction 结构,和transaction_stack类型相同;

struct binder_transaction {
    int debug_id;
    struct binder_work work;
    struct binder_thread *from;                                 ①
    struct binder_transaction *from_parent;                     ②
    struct binder_proc *to_proc;                                ③
    struct binder_thread *to_thread;                            ④
    struct binder_transaction *to_parent;                       ⑤
    unsigned need_reply:1;
    .....
};

①: 记录发送线程信息;
②: 记录发送线程的传输栈的父栈;
③: 记录接收进程的进程信息;
④: 记录接收线程的进程信息;
⑤: 记录接收进程的传输栈的父栈;

此时t变量中的主要信息如下:

transaction_starck
from	led_control_service'thread
to_proc	ServiceManager
to_thread	ServiceManager'thread

此时已经记录下了接收方的信息了,继续往下看(binder_transaction()函数内):

    if (reply) {
        BUG_ON(t->buffer->async_transaction != 0);
        binder_pop_transaction(target_thread, in_reply_to);
    } else if (!(t->flags & TF_ONE_WAY)) {                     ①
        BUG_ON(t->buffer->async_transaction != 0);
        t->need_reply = 1;
        t->from_parent = thread->transaction_stack;            ②
        thread->transaction_stack = t;                         ③
    } else {
        BUG_ON(target_node == NULL);
        BUG_ON(t->buffer->async_transaction != 1);
        if (target_node->has_async_transaction) {
            target_list = &target_node->async_todo;
            target_wait = NULL;
        } else
            target_node->has_async_transaction = 1;
    }

①: 判断是否需要回复;
②: 这里一个入栈操作将当前线程的传输压入,但是thread->transaction_stack此时为NULL,因为第一次执行前面没有赋值;
③: 接下给当前线程的传输栈赋值,;

此时led_control_service线程的传输栈信息如下:

transaction_starck
from	led_control_service'thread
to_proc	ServiceManager
to_thread	ServiceManager'thread
from_parent	NULL

2.2 BR_TRANSACTION

前面驱动讲解过,在led_contrl_server执行binder_transaction()后,ServiceManager的进程会被唤醒,则ServiceManager会从休眠中醒来继续执行binder_thread_read;

binder_thread_read有段代码如下:

    if (cmd == BR_TRANSACTION && !(t->flags & TF_ONE_WAY)) {
        t->to_parent = thread->transaction_stack;                ①
        t->to_thread = thread;                                   ②
        thread->transaction_stack = t;                           ③
    } else {
        t->buffer->transaction = NULL;
        kfree(t);
        binder_stats_deleted(BINDER_STAT_TRANSACTION);
    }

①: 将当前线程传输栈入栈;
②: 记录接收进程的信息,也就是自己本身,因为他本身是被唤醒的;
③: 当前线程传输栈记录下线程信息;

这里的t是从带处理事务的链表中取出来的,也就是前面led_control_service挂到ServiceManager的todo链表上;

则ServiceManager线程的传输栈的信息如下:

transaction_starck
from	led_control_service'thread
to_proc	ServiceManager
to_thread	ServiceManager'thread
from_parent	NULL
to_parent	NULL

到这里线程传输栈的数据来源和构造讲完;

三. transaction_stack的使用

ServiceManager的用户态在处理完注册信息后,调用BC_REPLAY命令回复注册结果给led_control_service,此时驱动中也是调用到binder_transaction();

binder_transaction代码片如下:

if (reply) {
        in_reply_to = thread->transaction_stack;                   ①
        ALOGD("%s:%d,%d %s in_reply_to = thread->transaction_stack\n", proc->tsk->comm, proc->pid, thread->pid, __FUNCTION__);
        if (in_reply_to == NULL) {
            binder_user_error("%d:%d got reply transaction with no transaction stack\n",
                      proc->pid, thread->pid);
            return_error = BR_FAILED_REPLY;
            goto err_empty_call_stack;
        }
        binder_set_nice(in_reply_to->saved_priority);
        if (in_reply_to->to_thread != thread) {                    ②
            return_error = BR_FAILED_REPLY;
            in_reply_to = NULL;
            goto err_bad_call_stack;
        }
        thread->transaction_stack = in_reply_to->to_parent;        ③
        target_thread = in_reply_to->from;                         ④
        if (target_thread == NULL) {
            return_error = BR_DEAD_REPLY;
            goto err_dead_binder;
        }
        if (target_thread->transaction_stack != in_reply_to) {
            return_error = BR_FAILED_REPLY;
            in_reply_to = NULL;
            target_thread = NULL;
            goto err_dead_binder;
        }
        target_proc = target_thread->proc;
    } else {

①: 用个临时变量记录下当前线程额传输栈信息;
②: 判断下接收的线程是否为自己本身,如果不是则出错,看迷糊的可以看下再2.2节;
③: 一次出栈操作,此时thread->transaction_stack值为NULL了;
④: 获取到目标线程,from中记录着发送线程led_contol_service的信息;

这里就通过ServiceManager在read的时候入栈的传送栈信息,获取到发送进程的信息,即回复进程的信息;

接着往下看:

    if (reply) {
        BUG_ON(t->buffer->async_transaction != 0);
        binder_pop_transaction(target_thread, in_reply_to);    ①
    } else if (!(t->flags & TF_ONE_WAY)) {
        BUG_ON(t->buffer->async_transaction != 0);
        t->need_reply = 1;
        t->from_parent = thread->transaction_stack;
        thread->transaction_stack = t;
    } else {
      ...
    }

①: 一个出栈操作;

此时led_control_service的传输栈也指向了父栈,即为空且清除了in_reply_to中from的信息;

上面那部ServiceManager进程已经知道此时要发送给谁,到此该问题就完美解答了;