从cgroup_init_early函数学习cgroup——初始化代码

前面分析了cgroup机制的框架及数据结构之间的关系，现在看代码就比较见简单了。 
代码部分 
在init/main.c的start_kernel函数中会调用cgroup_init_early函数进行cgroup的初始化。其实对cgroup的初始化分两次，分别是cgroup_init_early和cgroup_init。这么做的原因是，系统初始阶段需要使用一些 subsystem，先对这一部分进行初始化。 
来看一下cgroup_init_early

/**
 * cgroup_init_early - cgroup initialization at system boot
 *
 * Initialize cgroups at system boot, and initialize any
 * subsystems that request early init.
 */
int __init cgroup_init_early(void)
{
    struct cgroup_subsys *ss;
    int i;

    atomic_set(&init_css_set.refcount, 1);
    INIT_LIST_HEAD(&init_css_set.cgrp_links);
    INIT_LIST_HEAD(&init_css_set.tasks);
    INIT_HLIST_NODE(&init_css_set.hlist);
    css_set_count = 1;
    init_cgroup_root(&cgroup_dummy_root);
    cgroup_root_count = 1;
    RCU_INIT_POINTER(init_task.cgroups, &init_css_set);

    init_cgrp_cset_link.cset = &init_css_set;
    init_cgrp_cset_link.cgrp = cgroup_dummy_top;
    list_add(&init_cgrp_cset_link.cset_link, &cgroup_dummy_top->cset_links);
    list_add(&init_cgrp_cset_link.cgrp_link, &init_css_set.cgrp_links);

    /* at bootup time, we don't worry about modular subsystems */
    for_each_builtin_subsys(ss, i) {
        BUG_ON(!ss->name);
        BUG_ON(strlen(ss->name) > MAX_CGROUP_TYPE_NAMELEN);
        BUG_ON(!ss->css_alloc);
        BUG_ON(!ss->css_free);
        if (ss->subsys_id != i) {
            printk(KERN_ERR "cgroup: Subsys %s id == %d\n",
                   ss->name, ss->subsys_id);
            BUG();
        }

        if (ss->early_init)
            cgroup_init_subsys(ss);
    }
    return 0;
}

注释的说的很清楚了，initialize any subsystems that request early init. 
init_css_set是init进程使用的css_set结构体，先进行初始化： 
设置init_css_set的引用计数为1. 
初始化init_css_set.cgrp_links链表，该链表与cgrp_cset_link的cgrp_link相关联. 
初始化init_css_set.tasks链表，该链表与所有应用init_css_set的task相关联. 
初始化init_css_set.hlist链表，哈希表相关，没有深入研究. 
css_set_count用来记录系统css_set结构体变量的个数. 
初始化cgroup_dummy_root，定义为

static struct cgroupfs_root cgroup_dummy_root;
/*
 * A cgroupfs_root represents the root of a cgroup hierarchy, and may be
 * associated with a superblock to form an active hierarchy.  This is
 * internal to cgroup core.  Don't access directly from controllers.
 */
struct cgroupfs_root {
    struct super_block *sb;

    /* The bitmask of subsystems attached to this hierarchy */
    unsigned long subsys_mask;

    /* Unique id for this hierarchy. */
    int hierarchy_id;

    /* The root cgroup for this hierarchy */
    struct cgroup top_cgroup;

    /* Tracks how many cgroups are currently defined in hierarchy.*/
    int number_of_cgroups;

    /* A list running through the active hierarchies */
    struct list_head root_list;

    /* Hierarchy-specific flags */
    unsigned long flags;

    /* IDs for cgroups in this hierarchy */
    struct idr cgroup_idr;

    /* The path to use for release notifications. */
    char release_agent_path[PATH_MAX];

    /* The name for this hierarchy - may be empty */
    char name[MAX_CGROUP_ROOT_NAMELEN];
};

static void init_cgroup_root(struct cgroupfs_root *root)
{
    struct cgroup *cgrp = &root->top_cgroup;

    INIT_LIST_HEAD(&root->root_list);
    root->number_of_cgroups = 1;
    cgrp->root = root;
    RCU_INIT_POINTER(cgrp->name, &root_cgroup_name);
    init_cgroup_housekeeping(cgrp);
    idr_init(&root->cgroup_idr);
}

初始化cgroup_dummy_root的root_list链表 
将层次图中的cgroup个数设置为1 
将自身设置为层次图的顶端cgroup，即top_cgroup->root指向cgroup_dummy_root 
init_cgroup_housekeeping用于构建层次图的list变量和互斥变量 
初始化cgroup_dummy_root->cgroup_idr链表 
回到cgroup_init_early： 
设置cgroup_root_count = 1; 
接下来几行代码：

    init_cgrp_cset_link.cset = &init_css_set;
    init_cgrp_cset_link.cgrp = cgroup_dummy_top;
    list_add(&init_cgrp_cset_link.cset_link, &cgroup_dummy_top->cset_links);
    list_add(&init_cgrp_cset_link.cgrp_link, &init_css_set.cgrp_links);

内核定义

static struct cgrp_cset_link init_cgrp_cset_link;

/* dummy_top is a shorthand for the dummy hierarchy's top cgroup */
static struct cgroup * const cgroup_dummy_top = &cgroup_dummy_root.top_cgroup;

所以这里是将init进程的css_set(即init_css_set)与cgroup_dummy_root的top_cgroup(即cgroup_dummy_top)通过init_cgrp_cset_link相关联。 
继续看cgroup_init_early的剩余代码

    /* at bootup time, we don't worry about modular subsystems */
    for_each_builtin_subsys(ss, i) {
        BUG_ON(!ss->name);
        BUG_ON(strlen(ss->name) > MAX_CGROUP_TYPE_NAMELEN);
        BUG_ON(!ss->css_alloc);
        BUG_ON(!ss->css_free);
        if (ss->subsys_id != i) {
            printk(KERN_ERR "cgroup: Subsys %s id == %d\n",
                   ss->name, ss->subsys_id);
            BUG();
        }

        if (ss->early_init)
            cgroup_init_subsys(ss);
    }

这里应该是遍历所有builtin的subsystems并进行初始化，从注释不难看出，linux中subsystems可以分为builtin和modular两类。 
其中会做一些检查，检查子系统的默认信息是否有误，出现问题时停止初始化。检查之后调用cgroup_init_subsys进行初始化。 
这些buildtin的subsystem从何而来？

/**
 * for_each_builtin_subsys - iterate all built-in cgroup subsystems
 * @ss: the iteration cursor
 * @i: the index of @ss, CGROUP_BUILTIN_SUBSYS_COUNT after reaching the end
 *
 * Bulit-in subsystems are always present and iteration itself doesn't
 * require any synchronization.
 */
#define for_each_builtin_subsys(ss, i)                  \
    for ((i) = 0; (i) < CGROUP_BUILTIN_SUBSYS_COUNT &&      \
         (((ss) = cgroup_subsys[i]) || true); (i)++)

for_each_builtin_subsys是一个宏，展开并去掉BUG_ON信息后得到：

    for (i = 0; i < CGROUP_BUILTIN_SUBSYS_COUNT &&      
         (((ss) = cgroup_subsys[i]) || true); i++){
        if (ss->early_init)
            cgroup_init_subsys(ss);
    }

cgroup_subsys数组的定义在kernel/group.c

static struct cgroup_subsys *cgroup_subsys[CGROUP_SUBSYS_COUNT] = {
#include <linux/cgroup_subsys.h>
};

成员变量是#include

/*
 * List of cgroup subsystems.
 *
 * DO NOT ADD ANY SUBSYSTEM WITHOUT EXPLICIT ACKS FROM CGROUP MAINTAINERS.
 */
#if IS_SUBSYS_ENABLED(CONFIG_CPUSETS)
SUBSYS(cpuset)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_DEBUG)
SUBSYS(debug)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_SCHED)
SUBSYS(cpu_cgroup)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_CPUACCT)
SUBSYS(cpuacct)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_MEMCG)
SUBSYS(mem_cgroup)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_DEVICE)
SUBSYS(devices)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_FREEZER)
SUBSYS(freezer)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_NET_CLASSID)
SUBSYS(net_cls)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_BLK_CGROUP)
SUBSYS(blkio)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_PERF)
SUBSYS(perf)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_NET_PRIO)
SUBSYS(net_prio)
#endif

#if IS_SUBSYS_ENABLED(CONFIG_CGROUP_HUGETLB)
SUBSYS(hugetlb)
#endif
/*
 * DO NOT ADD ANY SUBSYSTEM WITHOUT EXPLICIT ACKS FROM CGROUP MAINTAINERS.
 */

一共有CGROUP_BUILTIN_SUBSYS_COUNT个subsystem,与inclue/linux/cgroup_subsys.h中subsystem中cgroup subsystems的数目相等。这个数值的定义用了一种很巧妙的方法，有兴趣的同学可以看一下。

接下来看cgroup_init_subsys

static void __init cgroup_init_subsys(struct cgroup_subsys *ss)
{
    struct cgroup_subsys_state *css;

    printk(KERN_INFO "Initializing cgroup subsys %s\n", ss->name);

    mutex_lock(&cgroup_mutex);

    /* init base cftset */
    cgroup_init_cftsets(ss);

    /* Create the top cgroup state for this subsystem */
    ss->root = &cgroup_dummy_root;
    css = ss->css_alloc(cgroup_css(cgroup_dummy_top, ss));
    /* We don't handle early failures gracefully */
    BUG_ON(IS_ERR(css));
    init_css(css, ss, cgroup_dummy_top);

    /* Update the init_css_set to contain a subsys
     * pointer to this state - since the subsystem is
     * newly registered, all tasks and hence the
     * init_css_set is in the subsystem's top cgroup. */
    init_css_set.subsys[ss->subsys_id] = css;

    need_forkexit_callback |= ss->fork || ss->exit;

    /* At system boot, before all subsystems have been
     * registered, no tasks have been forked, so we don't
     * need to invoke fork callbacks here. */
    BUG_ON(!list_empty(&init_task.tasks));

    BUG_ON(online_css(css));

    mutex_unlock(&cgroup_mutex);

    /* this function shouldn't be used with modular subsystems, since they
     * need to register a subsys_id, among other things */
    BUG_ON(ss->module);
}

这里主要是做一些初始化，由于时间问题，先不去深究了。 
这里我只是分析了一下cgroup的框架及初始化，其实还有很多值得研究的地方，比如subsys_id，比如cpu等cgroup是如何运作的，这写内容以后一定会补上。