Linux内核源代码情景分析-系统调用mknod
普通文件可以用open或者create创建，FIFO文件可以用pipe创建，mknod主要用于设备文件的创建。
在内核中，mknod是由sys_mknod实现的，代码如下：
asmlinkage long sys_mknod(const char * filename, int mode, dev_t dev) //比如filename为/tmp/server_socket，dev是设备号
{
int error = 0;
char * tmp;
struct dentry * dentry;
struct nameidata nd;
if (S_ISDIR(mode))
return -EPERM;
tmp = getname(filename);
if (IS_ERR(tmp))
return PTR_ERR(tmp);
if (path_init(tmp, LOOKUP_PARENT, &nd))//寻找父节点，这里就是/tmp节点
error = path_walk(tmp, &nd);
if (error)
goto out;
dentry = lookup_create(&nd, 0);//寻找/tmp/server_socket节点，返回该节点的dentry结构，但是dentry->d_inode为NULL
error = PTR_ERR(dentry);
if (!IS_ERR(dentry)) {
switch (mode & S_IFMT) {
case 0: case S_IFREG://普通文件
error = vfs_create(nd.dentry->d_inode,dentry,mode);
break;
case S_IFCHR: case S_IFBLK: case S_IFIFO: case S_IFSOCK://字符设备，块设备，管道，socket文件
error = vfs_mknod(nd.dentry->d_inode,mode,dev);//创建/tmp/server_socket节点的inode结构，并关联到文件系统中
break;
case S_IFDIR:
error = -EPERM;
break;
default:
error = -EINVAL;
}
dput(dentry);
}
up(&nd.dentry->d_inode->i_sem);
path_release(&nd);
out:
putname(tmp);
return error;
}
lookup_create，寻找/tmp/server_socket节点，代码如下：
static struct dentry *lookup_create(struct nameidata *nd, int is_dir)
{
struct dentry *dentry;
down(&nd->dentry->d_inode->i_sem);
dentry = ERR_PTR(-EEXIST);
if (nd->last_type != LAST_NORM)
goto fail;
dentry = lookup_hash(&nd->last, nd->dentry);//nd->last是server_socket
if (IS_ERR(dentry))
goto fail;
if (!is_dir && nd->last.name[nd->last.len] && !dentry->d_inode)
goto enoent;
return dentry;
enoent:
dput(dentry);
dentry = ERR_PTR(-ENOENT);
fail:
return dentry;
}
struct dentry * lookup_hash(struct qstr *name, struct dentry * base)//name为server_socket，base为父节点/tmp的dentry结构
{
struct dentry * dentry;
struct inode *inode;
int err;
inode = base->d_inode;//父节点/tmp的i节点
err = permission(inode, MAY_EXEC);
dentry = ERR_PTR(err);
if (err)
goto out;
/*
* See if the low-level filesystem might want
* to use its own hash..
*/
if (base->d_op && base->d_op->d_hash) {
err = base->d_op->d_hash(base, name);
dentry = ERR_PTR(err);
if (err < 0)
goto out;
}
dentry = cached_lookup(base, name, 0);
if (!dentry) {
struct dentry *new = d_alloc(base, name);//创建/tmp/server_socket节点的dentry结构
dentry = ERR_PTR(-ENOMEM);
if (!new)
goto out;
lock_kernel();
dentry = inode->i_op->lookup(inode, new);//dentry为NULL
unlock_kernel();
if (!dentry)
dentry = new;//刚刚创建的new赋值给dentry，但是dentry->d_inode为NULL
else
dput(new);
}
out:
return dentry;
} vfs_mknod，创建/tmp/server_socket节点的inode结构，并关联到文件系统中，代码如下：
int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)//dir为/tmp父节点的inode结构,dentry为/tmp/server_socket节点的dentry结构
{
int error = -EPERM;
mode &= ~current->fs->umask;
down(&dir->i_zombie);
if ((S_ISCHR(mode) || S_ISBLK(mode)) && !capable(CAP_MKNOD))//检验当前进程是否允许创建设备节点，此项检验仅用于待创建节点为设备节点时
goto exit_lock;
error = may_create(dir, dentry);
if (error)
goto exit_lock;
error = -EPERM;
if (!dir->i_op || !dir->i_op->mknod)
goto exit_lock;
DQUOT_INIT(dir);
lock_kernel();
error = dir->i_op->mknod(dir, dentry, mode, dev);//对于Ext2，这个函数是ext2_mknod
unlock_kernel();
exit_lock:
up(&dir->i_zombie);
if (!error)
inode_dir_notify(dir, DN_CREATE);
return error;
}
may_create，检查目标节点的inode结构是否存在。
static inline int may_create(struct inode *dir, struct dentry *child) {
if (child->d_inode)//也就是检查d_inode是否为NULL
return -EEXIST;
if (IS_DEADDIR(dir))
return -ENOENT;
return permission(dir,MAY_WRITE | MAY_EXEC);
}
对于Ext2，dir->i_op->mknod是ext2_mknod，代码如下：
static int ext2_mknod (struct inode * dir, struct dentry *dentry, int mode, int rdev)
{
struct inode * inode = ext2_new_inode (dir, mode);//分配了一个inode结构
int err = PTR_ERR(inode);
if (IS_ERR(inode))
return err;
inode->i_uid = current->fsuid;
init_special_inode(inode, mode, rdev);
err = ext2_add_entry (dir, dentry->d_name.name, dentry->d_name.len,
inode);//inode关联到文件系统中，也就是通过父节点inode结构，能够找到新创建的子节点的inode结构
if (err)
goto out_no_entry;
mark_inode_dirty(inode);//新创建的inode结构设置成"脏"
d_instantiate(dentry, inode);//将新创建的inode结构与dentry结构相关联
return 0;
out_no_entry:
inode->i_nlink--;
mark_inode_dirty(inode);
iput(inode);
return err;
}
void init_special_inode(struct inode *inode, umode_t mode, int rdev)
{
inode->i_mode = mode;
if (S_ISCHR(mode)) {//字符设备
inode->i_fop = &def_chr_fops;
inode->i_rdev = to_kdev_t(rdev);
} else if (S_ISBLK(mode)) {//块设备
inode->i_fop = &def_blk_fops;
inode->i_rdev = to_kdev_t(rdev);
inode->i_bdev = bdget(rdev);
} else if (S_ISFIFO(mode))//FIFO设备
inode->i_fop = &def_fifo_fops;
else if (S_ISSOCK(mode))//socket设备
inode->i_fop = &bad_sock_fops;
else
printk(KERN_DEBUG "init_special_inode: bogus imode (%o)\n", mode);
}
由于新创建的inode结构设置成了“脏”，内核在"同步"内存中的inode结构与磁盘上的索引节点的时候，就会将这个inode结构的内容写到磁盘上分配给这个文件的索引节点，即ext2_inode数据结构中。由于ext2_inode结构中并不存在i_rdev这么个成分，而对于设备文件却又不需要使用i_block[]数组，所以就挪用其i_block[0]来保存设备号。要了解这一点，主要看ext2_update_inode代码中一个片段：
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode))//FIFO设备和SOCKET设备没有设备号 raw_inode->i_block[0] = cpu_to_le32(kdev_t_to_nr(inode->i_rdev)); else for (block = 0; block < EXT2_N_BLOCKS; block++) raw_inode->i_block[block] = inode->u.ext2_i.i_data[block];
for (block = 0; block < EXT2_N_BLOCKS; block++)
inode->u.ext2_i.i_data[block] = raw_inode->i_block[block];
if (inode->i_ino == EXT2_ACL_IDX_INO ||
inode->i_ino == EXT2_ACL_DATA_INO)
/* Nothing to do */ ;
else if (S_ISREG(inode->i_mode)) {
} else if (S_ISDIR(inode->i_mode)) {
} else if (S_ISLNK(inode->i_mode)) {
} else
init_special_inode(inode, inode->i_mode,
le32_to_cpu(raw_inode->i_block[0]));
我们回过头想一想，在Linux内核源代码情景分析-文件系统的安装，/dev/sdb1，就是通过mknod建立的。
还有在Linux内核源代码情景分析-基于socket的进程间通信，/tmp/server_socket，也是通过vfs_mknod创建的。