devtmpfs文件系统创建设备节点

一、devtmpfs概述
1.devtmpfs 的功用是在 Linux 核心 启动早期建立一个初步的 /dev，令一般启动程序不用等待 udev，缩短 GNU/Linux 的开机时间。

2.重要解释
Devtmpfs lets the kernel create a tmpfs very early at kernel initialization, before any driver core device is registered. Every device with a major/minor will have a device node created in this tmpfs instance. After the rootfs is mounted by the kernel, the populated tmpfs is mounted at /dev. In initramfs, it can be moved to the manually mounted root filesystem before /sbin/init is executed.

3.menuconfig 中加入devtmpfs支持
make menuconfig-->Device Drivers-->Generic Driver Options
Maintain a devtmpfs filesystem to mount at /dev
Automount devtmpfs at /dev, after the kernel mounted the rootfs

4.df -T显示devtmpfs
文件系统    类型   1K-块      已用      可用 已用% 挂载点
/dev/sda1     ext4    31621016  14985712  15029008  50% /
none      devtmpfs      399552       276    399276   1% /dev
none         tmpfs      403804        24    403780   1% /dev/shm
none         tmpfs      403804       108    403696   1% /var/run
none         tmpfs      403804         0    403804   0% /var/lock
none         tmpfs      403804         0    403804   0% /lib/init/rw
.host:/ vmhgfs    67151668  54038400  13113268  81% /mnt/hgfs
/dev/loop0    ext2       16119      8528      6772  56% /mnt/loop

二、devtmpfs文件系统初始化
void __init driver_init(void)
{
    /* These are the core pieces */
    devtmpfs_init();//devtmpfs文件系统初始化
    devices_init();
    buses_init();
    classes_init();
    firmware_init();
    hypervisor_init();
    platform_bus_init();
    system_bus_init();
    cpu_dev_init();
    memory_dev_init();
}

static struct file_system_type dev_fs_type = {
    .name = "devtmpfs",
    .mount = dev_mount,
    .kill_sb = kill_litter_super,
};

int __init devtmpfs_init(void)
{
    int err = register_filesystem(&dev_fs_type);//注册dev_fs_type文件系统，即将dev_fs_type添加到内核全局总链表中file_systems
    if (err) {
        printk(KERN_ERR "devtmpfs: unable to register devtmpfs ""type %i\n", err);
        return err;
    }

    thread = kthread_run(devtmpfsd, &err, "kdevtmpfs");//创建并启动一个内核线程devtmpfsd
    if (!IS_ERR(thread)) {
        wait_for_completion(&setup_done);//进行一个不可打断的等待,允许一个线程告诉另一个线程工作已经完成
    } else {
        err = PTR_ERR(thread);
        thread = NULL;
    }

    if (err) {
        printk(KERN_ERR "devtmpfs: unable to create devtmpfs %i\n", err);
        unregister_filesystem(&dev_fs_type);
        return err;
    }

    printk(KERN_INFO "devtmpfs: initialized\n");
    return 0;
}

//请求创建设备节点的请求队列req结构
static struct req {
    struct req *next;
    struct completion done;
    int err;
    const char *name;
    umode_t mode;//0代表删除
    struct device *dev;
} *requests;

//内核线程devtmpfsd
static int devtmpfsd(void *p)
{
    char options[] = "mode=0755";
    int *err = p;

    *err = sys_unshare(CLONE_NEWNS);
    if (*err)
        goto out;

    //挂载devtmpfs文件系统
    //devtmpfs是待安装设备的路径名，“/”是安装点路径名，”devtmpfs“表示文件系统类型，MS_SILENT=32768，即0x8000
    *err = sys_mount("devtmpfs", "/", "devtmpfs", MS_SILENT, options);
    if (*err)
        goto out;
    sys_chdir("/.."); //将进程的当前工作目录(pwd)设定为devtmpfs文件系统的根目录/* will traverse into overmounted root */
    sys_chroot(".");
    complete(&setup_done);//允许一个线程告诉另一个线程工作已经完成
    while (1) {
        spin_lock(&req_lock);
        while (requests) {//扫描请求链表，每当要创建一个设备节点时，都需要向requests链表中添加请求
            struct req *req = requests;//赋值给临时req
            requests = NULL;//清空
            spin_unlock(&req_lock);
            while (req) {//遍历刚才requests的请求链表
                struct req *next = req->next;
                req->err = handle(req->name, req->mode, req->dev);//对链表中的每一个请求调用handle函数
                complete(&req->done);
                req = next;
            }
            spin_lock(&req_lock);
        }
        __set_current_state(TASK_INTERRUPTIBLE);//设置为睡眠状态
        spin_unlock(&req_lock);
        schedule();//系统切换
    }
    return 0;
out:
    complete(&setup_done);
    return *err;
}

static int handle(const char *name, umode_t mode, struct device *dev)
{
    if (mode)
        return handle_create(name, mode, dev);
    else
        return handle_remove(name, dev);
}

static int handle_create(const char *nodename, umode_t mode, struct device *dev)
{
    struct dentry *dentry;
    struct path path;
    int err;

    //查找节点名称的路径以及返回节点对应的父目录dentry结构，即在此目录下创建一个设备节点，即是/dev目录对应的dentry结构
    dentry = kern_path_create(AT_FDCWD, nodename, &path, 0);
    if (dentry == ERR_PTR(-ENOENT)) {
        create_path(nodename);
        dentry = kern_path_create(AT_FDCWD, nodename, &path, 0);
    }
    if (IS_ERR(dentry))
        return PTR_ERR(dentry);

    //创建设备节点
    err = vfs_mknod(path.dentry->d_inode,dentry, mode, dev->devt);
    if (!err) {
        struct iattr newattrs;
        newattrs.ia_mode = mode;/* fixup possibly umasked mode */
        newattrs.ia_valid = ATTR_MODE;
        mutex_lock(&dentry->d_inode->i_mutex);
        notify_change(dentry, &newattrs);
        mutex_unlock(&dentry->d_inode->i_mutex);
        dentry->d_inode->i_private = &thread;/* mark as kernel-created inode */
    }
    done_path_create(&path, dentry);//与前边kern_path_create对应，减少path和dentry的计数等
    return err;
}

int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev)
{
    int error = may_create(dir, dentry);//检查是否可以创建设备文件节点

    if (error)
        return error;

    //必须是字符设备或者块设备，且具有创建节点的权限
    if ((S_ISCHR(mode) || S_ISBLK(mode)) && !capable(CAP_MKNOD))
        return -EPERM;

    if (!dir->i_op->mknod)
        return -EPERM;

    error = devcgroup_inode_mknod(mode, dev);
    if (error)
        return error;

    error = security_inode_mknod(dir, dentry, mode, dev);
    if (error)
        return error;

    //调用具体文件系统的mknod()函数
    //mount时调用shmem_fill_super()-->shmem_get_inode()分配inode节点时做出的初始化
    /*那么在shmem_get_inode中
        caseS_IFDIR:
        inc_nlink(inode);
        inode->i_size= 2 * BOGO_DIRENT_SIZE;
        inode->i_op= &shmem_dir_inode_operations;
        inode->i_fop= &simple_dir_operations;
        由于mountpoint是dev这个目录，所以dev对应的inode的i_op就是shmem_dir_inode_operations。
        staticconst struct inode_operations shmem_dir_inode_operations = {
            #ifdefCONFIG_TMPFS
            .create =shmem_create,
            .lookup =simple_lookup,
            .link =shmem_link,
            .unlink =shmem_unlink,
            .symlink =shmem_symlink,
            .mkdir =shmem_mkdir,
            .rmdir =shmem_rmdir,
            .mknod =shmem_mknod,
            .rename =shmem_rename,
            #endif
            #ifdefCONFIG_TMPFS_POSIX_ACL
            .setattr =shmem_notify_change,
            .setxattr =generic_setxattr,
            .getxattr =generic_getxattr,
            .listxattr =generic_listxattr,
            .removexattr =generic_removexattr,
            .check_acl =generic_check_acl,
            #endif
            };
        */
    error = dir->i_op->mknod(dir, dentry, mode, dev);//所以这里调用的就是shmem_mknod
    if (!error)
        fsnotify_create(dir, dentry);
    return error;
}

shmem_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev)
{
    struct inode *inode;
    int error = -ENOSPC;

    inode = shmem_get_inode(dir->i_sb, dir, mode, dev, VM_NORESERVE);//获得一个要创建的设备节点的inode，并初始化
    if (inode) {
        error = security_inode_init_security(inode, dir,&dentry->d_name,shmem_initxattrs, NULL);
        if (error) {
            if (error != -EOPNOTSUPP) {
                iput(inode);
                return error;
            }
        }
#ifdef CONFIG_TMPFS_POSIX_ACL
        error = generic_acl_init(inode, dir);
        if (error) {
            iput(inode);
            return error;
        }
#else
        error = 0;
#endif
        dir->i_size += BOGO_DIRENT_SIZE;
        dir->i_ctime = dir->i_mtime = CURRENT_TIME;
        d_instantiate(dentry, inode);//与dentry建立关,此时就可以在/dev下看到这个字符设备节点了
        dget(dentry); //递减dentry的计数
    }
    return error;
}

三、文件系统的mount
内核主要是通过kernel_init调用prepare_namespace()函数执行安装实际根文件系统的操作：
void __init prepare_namespace(void)
{
    int is_floppy;

    if (root_delay) {
        printk(KERN_INFO "Waiting %dsec before mounting root device...\n",
               root_delay);
        ssleep(root_delay);
    }
    wait_for_device_probe();

    md_run_setup();
    /* 把root_device_name变量置为从启动参数“root”中获取的设备文件名。
  * 同样，把ROOT_DEV变量置为同一设备文件的主设备号和次设备号。*/
    if (saved_root_name[0]) {
        root_device_name = saved_root_name;
        if (!strncmp(root_device_name, "mtd", 3) ||
            !strncmp(root_device_name, "ubi", 3)) {
            mount_block_root(root_device_name, root_mountflags);
            goto out;
        }
        ROOT_DEV = name_to_dev_t(root_device_name);//转换为设备号/dev/mtdblock2.
        if (strncmp(root_device_name, "/dev/", 5) == 0)
            root_device_name += 5;
    }

    if (initrd_load())
        goto out;

    /* wait for any asynchronous scanning to complete */
    if ((ROOT_DEV == 0) && root_wait) {
        printk(KERN_INFO "Waiting for root device %s...\n",
            saved_root_name);
        while (driver_probe_done() != 0 ||
            (ROOT_DEV = name_to_dev_t(saved_root_name)) == 0)
            msleep(100);
        async_synchronize_full();
    }

    is_floppy = MAJOR(ROOT_DEV) == FLOPPY_MAJOR;

    if (is_floppy && rd_doload && rd_load_disk(0))
        ROOT_DEV = Root_RAM0;

    mount_root();
out:
    devtmpfs_mount("dev");//挂载devtmpfs文件系统
    sys_mount(".", "/", NULL, MS_MOVE, NULL); /* 移动rootfs文件系统根目录上的已安装文件系统的安装点。 */
    sys_chroot(".");
}

int devtmpfs_mount(const char *mntdir)
{
    int err;

    if (!mount_dev)
        return 0;

    if (!thread)
        return 0;
    //将devtmpfs文件系统挂载到/dev目录下
    err = sys_mount("devtmpfs", (char *)mntdir, "devtmpfs", MS_SILENT, NULL);
    if (err)
        printk(KERN_INFO "devtmpfs: error mounting %i\n", err);
    else
        printk(KERN_INFO "devtmpfs: mounted\n");
    return err;
}


四、devtmpfs创建节点
系统在启动过程中，扫描到的设备会通过devtmpfs_create_node()函数来添加设备节点。
int devtmpfs_create_node(struct device *dev)
{
    const char *tmp = NULL;
    struct req req;

    if (!thread)
    return 0;

    req.mode = 0;
    req.name = device_get_devnode(dev, &req.mode, &tmp);//获得设备名
    if (!req.name)
        return -ENOMEM;

    if (req.mode == 0)
        req.mode = 0600;
    if (is_blockdev(dev))
        req.mode |= S_IFBLK;//块设备
    else
        req.mode |= S_IFCHR;//字符设备

    req.dev = dev;

    init_completion(&req.done);

    spin_lock(&req_lock);
    req.next = requests;//请求添加到requests链表
    requests = &req;
    spin_unlock(&req_lock);

    wake_up_process(thread);//唤醒内核线程devtmpfsd添加设备节点
    wait_for_completion(&req.done);

    kfree(tmp);

    return req.err;
}

const char *device_get_devnode(struct device *dev,umode_t *mode, const char **tmp)
{
    char *s;

    *tmp = NULL;

    /* the device type may provide a specific name */
    if (dev->type && dev->type->devnode)
        *tmp = dev->type->devnode(dev, mode);
    if (*tmp)
        return *tmp;

    /* the class may provide a specific name */
    if (dev->class && dev->class->devnode)
        *tmp = dev->class->devnode(dev, mode);
    if (*tmp)
        return *tmp;

    /* return name without allocation, tmp == NULL */
    if (strchr(dev_name(dev), '!') == NULL)
        return dev_name(dev);

    /* replace '!' in the name with '/' */
    *tmp = kstrdup(dev_name(dev), GFP_KERNEL);
    if (!*tmp)
        return NULL;
    while ((s = strchr(*tmp, '!')))
        s[0] = '/';
    return *tmp;
}