super.c是ext4裏非常重要的文件，掛載時候運行的代碼基本上都在super.c裏邊，如果super.c有一點代碼上的bug的話，那麼一個塊設備就不能被以ext4文件系統掛載。我在比較重要的地方都加了註釋講解，如果有認爲我沒寫詳細或者寫錯的朋友們歡迎指出。
/*
* linux/fs/ext4/super.c
*
* Copyright (C) 1992, 1993, 1994, 1995
* Remy Card ([email protected])
* Laboratoire MASI - Institut Blaise Pascal
* Universite Pierre et Marie Curie (Paris VI)
* 好熟悉的人，我記得當初看ext2的時候作者也是這個人，巴黎第五大學的god
* from
*
* linux/fs/minix/inode.c
*
* Copyright (C) 1991, 1992 Linus Torvalds
*
* Big-endian to little-endian byte-swapping/bitmaps by
* David S. Miller ([email protected]), 1995
*/
#include <linux/module.h>
#include <linux/string.h>
#include <linux/fs.h>
#include <linux/time.h>
#include <linux/jbd2.h>
#include <linux/ext4_fs.h>
#include <linux/ext4_jbd2.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/parser.h>
#include <linux/smp_lock.h>
#include <linux/buffer_head.h>
#include <linux/vfs.h>
#include <linux/random.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/quotaops.h>
#include <linux/seq_file.h>
#include <asm/uaccess.h>
#include "xattr.h"
#include "acl.h"
#include "namei.h"
/* super.c裏會使用到的函數的聲明 */
static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
unsigned long journal_devnum);
static int ext4_create_journal(struct super_block *, struct ext4_super_block *,
unsigned int);
static void ext4_commit_super (struct super_block * sb,
struct ext4_super_block * es,
int sync);
static void ext4_mark_recovery_complete(struct super_block * sb,
struct ext4_super_block * es);
static void ext4_clear_journal_err(struct super_block * sb,
struct ext4_super_block * es);
static int ext4_sync_fs(struct super_block *sb, int wait);
static const char *ext4_decode_error(struct super_block * sb, int errno,
char nbuf[16]);
static int ext4_remount (struct super_block * sb, int * flags, char * data);
static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf);
static void ext4_unlockfs(struct super_block *sb);
static void ext4_write_super (struct super_block * sb);
static void ext4_write_super_lockfs(struct super_block *sb);
/* 根據ext4的超級塊的塊號以及對應組的組描述符獲得數據塊位圖的所在塊號 */
ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
struct ext4_group_desc *bg)
{
/* ext4是用的小端字節序，所以需要先轉換成本機的cpu字節序然後才能和內存裏的數進行比較加減等操作，下同。 */
/* 組描述符結構體裏的bg_block_bitmap字段表示數據塊所在的位置，EXT4_DESC_SIZE這個宏返回組描述符的大小，如果大於EXT4_MIN_DESC_SIZE_64BIT，就是大於64位的最小大小，說明是用64位來表示的，就把高32位也或上，這主要是爲了應對有的文件系統塊的數目太多32位已經表述不下來的情況 */
return le32_to_cpu(bg->bg_block_bitmap) |
(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
(ext4_fsblk_t)le32_to_cpu(bg->bg_block_bitmap_hi) << 32 : 0);
}
/* 根據ext4的超級塊的塊號以及對應組的組描述符獲得inode位圖的所在塊號 */
ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb,
struct ext4_group_desc *bg)
{
/* 組描述符結構體裏的bg_inode_bitmap字段表示存放inode的塊所在的位置，EXT4_DESC_SIZE這個宏返回組描述符的大小，如果大於EXT4_MIN_DESC_SIZE_64BIT，就是大於64位的最小大小，說明是用64位來表示的，就把高32位也或上，這主要是爲了應對有的文件系統塊的數目太多32位已經表述不下來的情況 */
return le32_to_cpu(bg->bg_inode_bitmap) |
(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
(ext4_fsblk_t)le32_to_cpu(bg->bg_inode_bitmap_hi) << 32 : 0);
}
/* 根據ext4的超級塊的塊號以及對應組的組描述符獲得inode table的開始塊號 */
ext4_fsblk_t ext4_inode_table(struct super_block *sb,
struct ext4_group_desc *bg)
{
/* 組描述符結構體裏的bg_inode_table字段表示inode table的開始塊所在的位置，EXT4_DESC_SIZE這個宏返回組描述符的大小，如果大於EXT4_MIN_DESC_SIZE_64BIT，就是大於64位的最小大小，說明是用64位來表示的，就把高32位也或上，這主要是爲了應對有的文件系統塊的數目太多32位已經表述不下來的情況 */
return le32_to_cpu(bg->bg_inode_table) |
(EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ?
(ext4_fsblk_t)le32_to_cpu(bg->bg_inode_table_hi) << 32 : 0);
}
/* 對塊組描述符結構體的bg_block_bitmap字段賦值爲blk */
void ext4_block_bitmap_set(struct super_block *sb,
struct ext4_group_desc *bg, ext4_fsblk_t blk)
{
/* 把blk轉換成cpu字節序賦值給塊組描述符的bg_block_bitmap，如果文件系統支持64位的，就把bg_block_bitmap_hi也就是高32位也賦值 */
bg->bg_block_bitmap = cpu_to_le32((u32)blk);
if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
bg->bg_block_bitmap_hi = cpu_to_le32(blk >> 32);
}
/* 對塊組描述符結構體的bg_inode_bitmap字段賦值爲blk，就是賦值inode位圖所在塊的賦值 */
void ext4_inode_bitmap_set(struct super_block *sb,
struct ext4_group_desc *bg, ext4_fsblk_t blk)
{
/* 把blk轉換成cpu字節序賦值給塊組描述符的bg_inode_bitmap，如果文件系統支持64位的，就把bg_inode_bitmap_hi也就是高32位也賦值 */
bg->bg_inode_bitmap = cpu_to_le32((u32)blk);
if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
bg->bg_inode_bitmap_hi = cpu_to_le32(blk >> 32);
}
/* 對塊組描述符結構體的bg_inode_table字段賦值爲blk，就是inode table開始塊的賦值 */
void ext4_inode_table_set(struct super_block *sb,
struct ext4_group_desc *bg, ext4_fsblk_t blk)
{
/* 把blk轉換成cpu字節序賦值給塊組描述符的bg_inode_bitmap，如果文件系統支持64位的，就把bg_inode_bitmap_hi也就是高32位也賦值 */
bg->bg_inode_table = cpu_to_le32((u32)blk);
if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT)
bg->bg_inode_table_hi = cpu_to_le32(blk >> 32);
}
/*
* ext4對於jbd2_journal_start/end的封裝，參數nblocks表示這次日誌記錄可能修改的塊的數目
* journal其實是內核裏的另外一個模塊實現的，ext4這裏僅僅是調用了journal模塊的接口，在2.6.19內核版本里使用的是journal2，也就是jbd2。
* 對於ext4封裝的日誌函數而言，我們唯一需要做的特殊的事情就是確保journal_end 的調用函數的結果在superblock裏，需要標記爲髒所以sync()就在合適的時候會調用文件系統的write_super回調函數。
*/
handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks)
{
journal_t *journal;
/* 如果文件系統是制度掛載的話，不支持日誌，一般只讀掛載都是出錯了 */
if (sb->s_flags & MS_RDONLY)
return ERR_PTR(-EROFS);
/* 這裏如果日誌aborted的，我們需要報錯，如果日誌沒有aborted狀態，我們直接調用jbd2模塊的函數jbd2_journal_start 來貨的日誌句柄 */
journal = EXT4_SB(sb)->s_journal;
if (is_journal_aborted(journal)) {
ext4_abort(sb, __FUNCTION__,
"Detected aborted journal");
return ERR_PTR(-EROFS);
}
return jbd2_journal_start(journal, nblocks);
}
/* 對於jbd2_journal_stop的封裝，做一些基本的檢查以後就調用jbd2_journal_stop函數。
* 我們唯一需要做的特殊的事情就是確保journal_end 的調用函數的結果在superblock裏，需要標記爲髒所以sync()就在合適的時候會調用文件系統的write_super回調函數。
*/
int __ext4_journal_stop(const char *where, handle_t *handle)
{
struct super_block *sb;
int err;
int rc;
sb = handle->h_transaction->t_journal->j_private;
err = handle->h_err;
rc = jbd2_journal_stop(handle);
if (!err)
err = rc;
if (err)
__ext4_std_error(sb, where, err);
return err;
}
/* 中途停止日誌首先打印出錯誤信息，然後調用jbd2的jbd2_journal_abort_handle函數來停止日誌 */
void ext4_journal_abort_handle(const char *caller, const char *err_fn,
struct buffer_head *bh, handle_t *handle, int err)
{
/* 從err獲得錯誤信息 */
char nbuf[16];
const char *errstr = ext4_decode_error(NULL, err, nbuf);
/* befferhead記錄錯誤信息 */
if (bh)
BUFFER_TRACE(bh, "abort");
/* 記錄錯誤值在句柄中 */
if (!handle->h_err)
handle->h_err = err;
/* 如果句柄所在的事務已經被終止了，我們就不必再次終止了 */
if (is_handle_aborted(handle))
return;
printk(KERN_ERR "%s: aborting transaction: %s in %s\n",
caller, errstr, err_fn);
jbd2_journal_abort_handle(handle);
}
/* 當ext4報錯誤的時候的處理函數，比如一些一致性錯誤或者是讀寫的IO錯誤。
* 在ext2上，我們把一些錯誤狀態字存儲在超級塊裏。但是在ext4，這不可能，因爲我們可能有寫入順序限制在超級塊上，這個限制將會阻止我們寫入，考慮到日誌將要終止的情況，我們不能依賴以後。
* 我們使用jbd2_journal_abort()獲得的錯誤碼來記錄錯誤，在恢復的時候日誌將會報錯誤，直到我們清除這個錯誤記錄。 */
static void ext4_handle_error(struct super_block *sb)
{
/* 首先從超級塊結構體獲得磁盤上存儲的ext4超級塊結構體 */
struct ext4_super_block *es = EXT4_SB(sb)->s_es;
/* 掛載狀態標記爲錯誤發生狀態 */
EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
/* 如果已經是隻讀掛載了，我們就不必繼續操作了 */
if (sb->s_flags & MS_RDONLY)
return;
/* 如果不是掛載選項是出錯以後繼續，那麼就停止一切事務 */
if (!test_opt (sb, ERRORS_CONT)) {
journal_t *journal = EXT4_SB(sb)->s_journal;
EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT;
if (journal)
jbd2_journal_abort(journal, -EIO);
}
/* 如果fs是標記未，如果有錯誤，重新掛載爲只讀的情況，我們就重新掛載爲只讀 */
if (test_opt (sb, ERRORS_RO)) {
printk (KERN_CRIT "Remounting filesystem read-only\n");
sb->s_flags |= MS_RDONLY;
}
/* 提交超級塊的修改 */
ext4_commit_super(sb, es, 1);
/* 如果掛載選項是出錯就panic，打印出panic信息 */
if (test_opt(sb, ERRORS_PANIC))
panic("EXT4-fs (device %s): panic forced after error\n",
sb->s_id);
}
/* 調用上邊的函數處理錯誤，在此之前打印出錯誤信息 */
void ext4_error (struct super_block * sb, const char * function,
const char * fmt, ...)
{
va_list args;
/* 先打印出錯誤信息 */
va_start(args, fmt);
printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function);
vprintk(fmt, args);
printk("\n");
va_end(args);
/* 然後處理錯誤 */
ext4_handle_error(sb);
}
/* 根據錯誤編號返回相應的錯誤信息字符串 */
static const char *ext4_decode_error(struct super_block * sb, int errno,
char nbuf[16])
{
char *errstr = NULL;
switch (errno) {
/* IO讀寫錯誤 */
case -EIO:
errstr = "IO failure";
break;
/* 內存滿了 */
case -ENOMEM:
errstr = "Out of memory";
break;
/* 只讀的文件系統，如果日誌系統也崩潰了的話，就返回日誌終止的信息 */
case -EROFS:
if (!sb || EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT)
errstr = "Journal has aborted";
else
errstr = "Readonly filesystem";
break;
default:
/* 如果用戶傳入的錯誤碼是未知的我們就直接返回error:錯誤碼的值,但是如果用戶傳入的緩衝區都是空的，那就直接返回NULL */
if (nbuf) {
if (snprintf(nbuf, 16, "error %d", -errno) >= 0)
errstr = nbuf;
}
break;
}
return errstr;
}
/* __ext4_std_error函數自動的從日誌裏接受一些希望得到的錯誤碼，然後做出合理的反應. */
void __ext4_std_error (struct super_block * sb, const char * function,
int errno)
{
char nbuf[16];
const char *errstr;
/* 特殊情況，如果錯誤碼是EROFS,並且我忙呢又不在事務裏，就沒必要記錄這個錯誤 */
if (errno == -EROFS && journal_current_handle() == NULL &&
(sb->s_flags & MS_RDONLY))
return;
/* 解碼返回錯誤值 */
errstr = ext4_decode_error(sb, errno, nbuf);
printk (KERN_CRIT "EXT4-fs error (device %s) in %s: %s\n",
sb->s_id, function, errstr);
/* 處理錯誤 */
ext4_handle_error(sb);
}
/*
* ext4_abort是一個比ext4_error健壯很多的錯誤處理函數.這個函數可以用來處理一些不可恢復的錯誤，比如to或者一些嚴重的日誌錯誤
* 我們將會強制ext4進入ABORT|READONLY狀態，除非文件系統的掛載選項是要求我們立刻panic當遇到錯誤的時候，這時我們就直接panic
*/
void ext4_abort (struct super_block * sb, const char * function,
const char * fmt, ...)
{
/* 首先打印用戶傳入的錯誤信息 */
va_list args;
printk (KERN_CRIT "ext4_abort called.\n");
va_start(args, fmt);
printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function);
vprintk(fmt, args);
printk("\n");
va_end(args);
/* 如果掛載選項是出現錯誤就panic，打印panic信息 */
if (test_opt(sb, ERRORS_PANIC))
panic("EXT4-fs panic from previous error\n");
/* 已經是隻讀，就沒必要接下來的操作了 */
if (sb->s_flags & MS_RDONLY)
return;
/* 如果不是隻讀掛載的，我們需要在超級塊上記錄錯誤，然後只讀掛載，最後終止日誌 */
printk(KERN_CRIT "Remounting filesystem read-only\n");
EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
sb->s_flags |= MS_RDONLY;
EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT;
jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
}
/* 簡單的打印警告信息 */
void ext4_warning (struct super_block * sb, const char * function,
const char * fmt, ...)
{
va_list args;
va_start(args, fmt);
printk(KERN_WARNING "EXT4-fs warning (device %s): %s: ",
sb->s_id, function);
vprintk(fmt, args);
printk("\n");
va_end(args);
}
/* 動態更新到ext4的新版 */
void ext4_update_dynamic_rev(struct super_block *sb)
{
struct ext4_super_block *es = EXT4_SB(sb)->s_es;
/* 如果已經是新版，就不用更新了 */
if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV)
return;
/* 動態更新是一件很危險的事情，因爲要修改inode編號以及大小 */
ext4_warning(sb, __FUNCTION__,
"updating to rev %d because of new feature flag, "
"running e2fsck is recommended",
EXT4_DYNAMIC_REV);
/* 修改文件系統的第一個inode號碼，inode大小，版本號修改 */
es->s_first_ino = cpu_to_le32(EXT4_GOOD_OLD_FIRST_INO);
es->s_inode_size = cpu_to_le16(EXT4_GOOD_OLD_INODE_SIZE);
es->s_rev_level = cpu_to_le32(EXT4_DYNAMIC_REV);
/*superblock的其他字段應該是0，如果不是，說明他們正在被使用，不管他們就行了。交給e2fsck來清理不一致的狀態。 */
}
/*
* 打開外部的日誌塊設備，參數是設備號
*/
static struct block_device *ext4_blkdev_get(dev_t dev)
{
struct block_device *bdev;
char b[BDEVNAME_SIZE];
/* 調用其他層給的接口函數來貨的block device */
bdev = open_by_devnum(dev, FMODE_READ|FMODE_WRITE);
if (IS_ERR(bdev))
goto fail;
return bdev;
fail:
printk(KERN_ERR "EXT4: failed to open journal device %s: %ld\n",
__bdevname(dev, b), PTR_ERR(bdev));
return NULL;
}
/*
* 釋放日誌設備，也是調用其他層給的接口函數來釋放
*/
static int ext4_blkdev_put(struct block_device *bdev)
{
bd_release(bdev);
return blkdev_put(bdev);
}
/* 移除超級塊信息結構體裏的塊設備 */
static int ext4_blkdev_remove(struct ext4_sb_info *sbi)
{
struct block_device *bdev;
int ret = -ENODEV;
bdev = sbi->journal_bdev;
if (bdev) {
/* ext4_blkdev_put來釋放，然後賦值爲NULL */
ret = ext4_blkdev_put(bdev);
sbi->journal_bdev = NULL;
}
return ret;
}
/* 由list_head來獲得vfs_inode */
static inline struct inode *orphan_list_entry(struct list_head *l)
{
return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode;
}
/* 打印ext4_sb_info結構體上的s_orphan鏈表所有項 */
static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi)
{
struct list_head *l;
printk(KERN_ERR "sb orphan head is %d\n",
le32_to_cpu(sbi->s_es->s_last_orphan));
printk(KERN_ERR "sb_info orphan list:\n");
/* 遍歷所有項，對於每一個項打印 */
list_for_each(l, &sbi->s_orphan) {
struct inode *inode = orphan_list_entry(l);
printk(KERN_ERR " "
"inode %s:%lu at %p: mode %o, nlink %d, next %d\n",
inode->i_sb->s_id, inode->i_ino, inode,
inode->i_mode, inode->i_nlink,
NEXT_ORPHAN(inode));
}
}
/* 卸載文件系統的時候調用的函數 */
static void ext4_put_super (struct super_block * sb)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
struct ext4_super_block *es = sbi->s_es;
int i;
/* 釋放超級塊，以及超級塊附着的擴展屬性結構體，還有ext4的日誌結構體 */
ext4_ext_release(sb);
ext4_xattr_put_super(sb);
jbd2_journal_destroy(sbi->s_journal);
/*如果現在是隻讀掛載，說明現在在恢復，清理掉恢復中的標記，然後把超級塊的buffer head寫入磁盤設備*/
if (!(sb->s_flags & MS_RDONLY)) {
EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
es->s_state = cpu_to_le16(sbi->s_mount_state);
BUFFER_TRACE(sbi->s_sbh, "marking dirty");
mark_buffer_dirty(sbi->s_sbh);
ext4_commit_super(sb, es, 1);
}
/* 對於每一個組，減少對應的塊組描述符引用計數 */
for (i = 0; i < sbi->s_gdb_count; i++)
brelse(sbi->s_group_desc[i]);
/* 釋放塊組描述符佔用的內存 */
kfree(sbi->s_group_desc);
/* 釋放ext4_sb_info裏的單cpu變量 */
percpu_counter_destroy(&sbi->s_freeblocks_counter);
percpu_counter_destroy(&sbi->s_freeinodes_counter);
percpu_counter_destroy(&sbi->s_dirs_counter);
/* 釋放ext4_sb_info結構體 */
brelse(sbi->s_sbh);
#ifdef CONFIG_QUOTA
/* 如果配置了配額，我們還需要釋放所有的配額相關的資源 */
for (i = 0; i < MAXQUOTAS; i++)
kfree(sbi->s_qf_names[i]);
#endif
/* 調試打印信息 */
if (!list_empty(&sbi->s_orphan))
dump_orphan_list(sb, sbi);
J_ASSERT(list_empty(&sbi->s_orphan));
invalidate_bdev(sb->s_bdev, 0);
if (sbi->journal_bdev && sbi->journal_bdev != sb->s_bdev) {
/*
* 遇到了非法的日誌設備緩衝區
*/
sync_blockdev(sbi->journal_bdev);
invalidate_bdev(sbi->journal_bdev, 0);
ext4_blkdev_remove(sbi);
}
sb->s_fs_info = NULL;
kfree(sbi);
return;
}
/* inode緩存分配的地方 */
static kmem_cache_t *ext4_inode_cachep;
/*
* 分配inode的函數，因爲調用了內部的事務程序，所以使用GFP_NOFS標誌
*/
static struct inode *ext4_alloc_inode(struct super_block *sb)
{
struct ext4_inode_info *ei;
/* 分配一部分緩衝區 */
ei = kmem_cache_alloc(ext4_inode_cachep, SLAB_NOFS);
if (!ei)
return NULL;
/* 如果配置了權限控制列表，初始化這些字段 */
#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
ei->i_acl = EXT4_ACL_NOT_CACHED;
ei->i_default_acl = EXT4_ACL_NOT_CACHED;
#endif
/* 初始化一些inode的字段 */
ei->i_block_alloc_info = NULL;
ei->vfs_inode.i_version = 1;
memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache));
/* 最後返回分配的inode */
return &ei->vfs_inode;
}
/* 釋放inode回內存的緩衝區 */
static void ext4_destroy_inode(struct inode *inode)
{
kmem_cache_free(ext4_inode_cachep, EXT4_I(inode));
}
/* 初始化ext4_inode_info結構體 */
static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
{
struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
SLAB_CTOR_CONSTRUCTOR) {
INIT_LIST_HEAD(&ei->i_orphan);
#ifdef CONFIG_EXT4DEV_FS_XATTR
init_rwsem(&ei->xattr_sem);
#endif
mutex_init(&ei->truncate_mutex);
inode_init_once(&ei->vfs_inode);
}
}
/* 初始化inode緩存slab分配器，如果是啊比返回ENOMEM */
static int init_inodecache(void)
{
ext4_inode_cachep = kmem_cache_create("ext4_inode_cache",
sizeof(struct ext4_inode_info),
0, (SLAB_RECLAIM_ACCOUNT|
SLAB_MEM_SPREAD),
init_once, NULL);
if (ext4_inode_cachep == NULL)
return -ENOMEM;
return 0;
}
/* 銷燬inode緩存 */
static void destroy_inodecache(void)
{
kmem_cache_destroy(ext4_inode_cachep);
}
/* 清理inode結構體以及相關的資源 */
static void ext4_clear_inode(struct inode *inode)
{
struct ext4_block_alloc_info *rsv = EXT4_I(inode)->i_block_alloc_info;
/* 如果配置了posix標準的權限控制列表，就是放acl相關的資源 */
#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
if (EXT4_I(inode)->i_acl &&
EXT4_I(inode)->i_acl != EXT4_ACL_NOT_CACHED) {
posix_acl_release(EXT4_I(inode)->i_acl);
EXT4_I(inode)->i_acl = EXT4_ACL_NOT_CACHED;
}
if (EXT4_I(inode)->i_default_acl &&
EXT4_I(inode)->i_default_acl != EXT4_ACL_NOT_CACHED) {
posix_acl_release(EXT4_I(inode)->i_default_acl);
EXT4_I(inode)->i_default_acl = EXT4_ACL_NOT_CACHED;
}
#endif
/* 把inode保留的塊都釋放了，一些字段賦值爲NULL或釋放掉 */
ext4_discard_reservation(inode);
EXT4_I(inode)->i_block_alloc_info = NULL;
if (unlikely(rsv))
kfree(rsv);
}
/* 打印配額選項 */
static inline void ext4_show_quota_options(struct seq_file *seq, struct super_block *sb)
{
#if defined(CONFIG_QUOTA)
struct ext4_sb_info *sbi = EXT4_SB(sb);
if (sbi->s_jquota_fmt)
seq_printf(seq, ",jqfmt=%s",
(sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0");
if (sbi->s_qf_names[USRQUOTA])
seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]);
if (sbi->s_qf_names[GRPQUOTA])
seq_printf(seq, ",grpjquota=%s", sbi->s_qf_names[GRPQUOTA]);
if (sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA)
seq_puts(seq, ",usrquota");
if (sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA)
seq_puts(seq, ",grpquota");
#endif
}
/* 打印掛載時候的數據模式 */
static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
{
struct super_block *sb = vfs->mnt_sb;
if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA)
seq_puts(seq, ",data=journal");
else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
seq_puts(seq, ",data=ordered");
else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)
seq_puts(seq, ",data=writeback");
ext4_show_quota_options(seq, sb);
return 0;
}
/* 從ext4獲取一個dentry結構體，vobjp[0]是inode編號，vobjp[1]是版本號 */
static struct dentry *ext4_get_dentry(struct super_block *sb, void *vobjp)
{
__u32 *objp = vobjp;
unsigned long ino = objp[0];
__u32 generation = objp[1];
struct inode *inode;
struct dentry *result;
/* 首先檢查inode編號的合法性 */
if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
return ERR_PTR(-ESTALE);
if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
return ERR_PTR(-ESTALE);
/* 根據inode編號返回對應的inode結構體 */
inode = iget(sb, ino);
if (inode == NULL)
return ERR_PTR(-ENOMEM);
/* 說明inode已經被刪除了，我們返回對應的錯誤碼 */
if (is_bad_inode(inode) ||
(generation && inode->i_generation != generation)) {
iput(inode);
return ERR_PTR(-ESTALE);
}
/* 根據inode獲得dentry */
result = d_alloc_anon(inode);
if (!result) {
iput(inode);
return ERR_PTR(-ENOMEM);
}
return result;
}
#ifdef CONFIG_QUOTA
#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group")
#define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
/* 如果定義了CONFIG_QUOTA宏，就定義和配額相關的操作 */
static int ext4_dquot_initialize(struct inode *inode, int type);
static int ext4_dquot_drop(struct inode *inode);
static int ext4_write_dquot(struct dquot *dquot);
static int ext4_acquire_dquot(struct dquot *dquot);
static int ext4_release_dquot(struct dquot *dquot);
static int ext4_mark_dquot_dirty(struct dquot *dquot);
static int ext4_write_info(struct super_block *sb, int type);
static int ext4_quota_on(struct super_block *sb, int type, int format_id, char *path);
static int ext4_quota_on_mount(struct super_block *sb, int type);
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
size_t len, loff_t off);
static ssize_t ext4_quota_write(struct super_block *sb, int type,
const char *data, size_t len, loff_t off);
/* 配額操作結構體 */
static struct dquot_operations ext4_quota_operations = {
.initialize = ext4_dquot_initialize,
.drop = ext4_dquot_drop,
.alloc_space = dquot_alloc_space,
.alloc_inode = dquot_alloc_inode,
.free_space = dquot_free_space,
.free_inode = dquot_free_inode,
.transfer = dquot_transfer,
.write_dquot = ext4_write_dquot,
.acquire_dquot = ext4_acquire_dquot,
.release_dquot = ext4_release_dquot,
.mark_dirty = ext4_mark_dquot_dirty,
.write_info = ext4_write_info
};
/* 響應用戶傳來的配額相關的請求的函數結構體 */
static struct quotactl_ops ext4_qctl_operations = {
.quota_on = ext4_quota_on,
.quota_off = vfs_quota_off,
.quota_sync = vfs_quota_sync,
.get_info = vfs_get_dqinfo,
.set_info = vfs_set_dqinfo,
.get_dqblk = vfs_get_dqblk,
.set_dqblk = vfs_set_dqblk
};
#endif
/* ext4文件系統的超級塊的操作函數結構體 */
static struct super_operations ext4_sops = {
.alloc_inode = ext4_alloc_inode,
.destroy_inode = ext4_destroy_inode,
.read_inode = ext4_read_inode,
.write_inode = ext4_write_inode,
.dirty_inode = ext4_dirty_inode,
.delete_inode = ext4_delete_inode,
.put_super = ext4_put_super,
.write_super = ext4_write_super,
.sync_fs = ext4_sync_fs,
.write_super_lockfs = ext4_write_super_lockfs,
.unlockfs = ext4_unlockfs,
.statfs = ext4_statfs,
.remount_fs = ext4_remount,
.clear_inode = ext4_clear_inode,
.show_options = ext4_show_options,
#ifdef CONFIG_QUOTA
.quota_read = ext4_quota_read,
.quota_write = ext4_quota_write,
#endif
};
/* 提供給vfs層調用的函數 */
static struct export_operations ext4_export_ops = {
.get_parent = ext4_get_parent,
.get_dentry = ext4_get_dentry,
};
/* 文件系統掛載的所有選項，以枚舉變量的形式展現 */
enum {
Opt_bsd_df, Opt_minix_df, Opt_grpid, Opt_nogrpid,
Opt_resgid, Opt_resuid, Opt_sb, Opt_err_cont, Opt_err_panic, Opt_err_ro,
Opt_nouid32, Opt_nocheck, Opt_debug, Opt_oldalloc, Opt_orlov,
Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh,
Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev,
Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
Opt_grpquota, Opt_extents,
};
/* 掛載選項和字符串的對應關係 */
static match_table_t tokens = {
{Opt_bsd_df, "bsddf"},
{Opt_minix_df, "minixdf"},
{Opt_grpid, "grpid"},
{Opt_grpid, "bsdgroups"},
{Opt_nogrpid, "nogrpid"},
{Opt_nogrpid, "sysvgroups"},
{Opt_resgid, "resgid=%u"},
{Opt_resuid, "resuid=%u"},
{Opt_sb, "sb=%u"},
{Opt_err_cont, "errors=continue"},
{Opt_err_panic, "errors=panic"},
{Opt_err_ro, "errors=remount-ro"},
{Opt_nouid32, "nouid32"},
{Opt_nocheck, "nocheck"},
{Opt_nocheck, "check=none"},
{Opt_debug, "debug"},
{Opt_oldalloc, "oldalloc"},
{Opt_orlov, "orlov"},
{Opt_user_xattr, "user_xattr"},
{Opt_nouser_xattr, "nouser_xattr"},
{Opt_acl, "acl"},
{Opt_noacl, "noacl"},
{Opt_reservation, "reservation"},
{Opt_noreservation, "noreservation"},
{Opt_noload, "noload"},
{Opt_nobh, "nobh"},
{Opt_bh, "bh"},
{Opt_commit, "commit=%u"},
{Opt_journal_update, "journal=update"},
{Opt_journal_inum, "journal=%u"},
{Opt_journal_dev, "journal_dev=%u"},
{Opt_abort, "abort"},
{Opt_data_journal, "data=journal"},
{Opt_data_ordered, "data=ordered"},
{Opt_data_writeback, "data=writeback"},
{Opt_offusrjquota, "usrjquota="},
{Opt_usrjquota, "usrjquota=%s"},
{Opt_offgrpjquota, "grpjquota="},
{Opt_grpjquota, "grpjquota=%s"},
{Opt_jqfmt_vfsold, "jqfmt=vfsold"},
{Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
{Opt_grpquota, "grpquota"},
{Opt_noquota, "noquota"},
{Opt_quota, "quota"},
{Opt_usrquota, "usrquota"},
{Opt_barrier, "barrier=%u"},
{Opt_extents, "extents"},
{Opt_err, NULL},
{Opt_resize, "resize"},
};
/* data參數是掛載選項的字符串，這個函數的作用是返回超級塊所在的物理塊號，如果用戶指定了
超級塊的位置比如sb=xxx，就返回超級塊的位置，如果沒有指定，就返回1，說明在默認的位置 */
static ext4_fsblk_t get_sb_block(void **data)
{
ext4_fsblk_t sb_block;
/* 用char指向傳入的掛載選項字符串 */
char *options = (char *) *data;
/* 如果參數是NULL或者掛載選項沒有指定超級塊所在的位置，就說明超級塊在默認的位置 */
if (!options || strncmp(options, "sb=", 3) != 0)
return 1;
/* 運行到這裏就說明用戶指定了超級塊的位置 */
options += 3;
/*把字符串轉換成超級塊數位置 */
sb_block = simple_strtoul(options, &options, 0);
if (*options && *options != ',') {
printk("EXT4-fs: Invalid sb specification: %s\n",
(char *) *data);
return 1;
}
if (*options == ',')
options++;
*data = (void *) options;
return sb_block;
}
/* 把用戶傳入的字符串形式的掛載參數轉化成s_mount_opt上的對應位圖的表現形式 */
static int parse_options (char *options, struct super_block *sb,
unsigned int *inum, unsigned long *journal_devnum,
ext4_fsblk_t *n_blocks_count, int is_remount)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
char * p;
substring_t args[MAX_OPT_ARGS];
int data_opt = 0;
int option;
#ifdef CONFIG_QUOTA
int qtype;
char *qname;
#endif
/* 檢查字符串參數 */
if (!options)
return 1;
/* 逐個解析掛載選項，並在s_mount_opt上對相應位置位 */
while ((p = strsep (&options, ",")) != NULL) {
int token;
if (!*p)
continue;
/* 把字符串選項轉換成枚舉體裏的變量 */
token = match_token(p, tokens, args);
switch (token) {
/* 對於bsddf和minixdf都認爲是minix文件系統 */
case Opt_bsd_df:
clear_opt (sbi->s_mount_opt, MINIX_DF);
break;
case Opt_minix_df:
set_opt (sbi->s_mount_opt, MINIX_DF);
break;
/* 設置組有沒有默認的組ID，如果有就置位，沒有就清除對應位 */
case Opt_grpid:
set_opt (sbi->s_mount_opt, GRPID);
break;
/* 新創建的文件的組ID是創建者 */
case Opt_nogrpid:
clear_opt (sbi->s_mount_opt, GRPID);
break;
/* 設置可能是用保留塊的用戶ID */
case Opt_resuid:
if (match_int(&args[0], &option))
return 0;
sbi->s_resuid = option;
break;
/* 設置可能使用保留塊的組ID */
case Opt_resgid:
if (match_int(&args[0], &option))
return 0;
sbi->s_resgid = option;
break;
case Opt_sb:
/* 已經移動到了get_sb_block()函數中處理這裏不再處理 */
break;
/* 這三個宏表示出現錯誤的時候是繼續還是重新掛載成爲只讀，亦或是報PANIC，在相應位上置位或者清除位 */
case Opt_err_panic:
clear_opt (sbi->s_mount_opt, ERRORS_CONT);
clear_opt (sbi->s_mount_opt, ERRORS_RO);
set_opt (sbi->s_mount_opt, ERRORS_PANIC);
break;
case Opt_err_ro:
clear_opt (sbi->s_mount_opt, ERRORS_CONT);
clear_opt (sbi->s_mount_opt, ERRORS_PANIC);
set_opt (sbi->s_mount_opt, ERRORS_RO);
break;
case Opt_err_cont:
clear_opt (sbi->s_mount_opt, ERRORS_RO);
clear_opt (sbi->s_mount_opt, ERRORS_PANIC);
set_opt (sbi->s_mount_opt, ERRORS_CONT);
break;
/* 設置沒有默認文件uid，於是創建者就是文件的uid */
case Opt_nouid32:
set_opt (sbi->s_mount_opt, NO_UID32);
break;
/* 掛載的時候不會進行位圖的檢查 */
case Opt_nocheck:
clear_opt (sbi->s_mount_opt, CHECK);
break;
/* 調試信息會不會在syslog中打印出來 */
case Opt_debug:
set_opt (sbi->s_mount_opt, DEBUG);
break;
/* 不使用Orlov塊分配器 */
case Opt_oldalloc:
set_opt (sbi->s_mount_opt, OLDALLOC);
break;
/* 使用Orlov塊分配器 */
case Opt_orlov:
clear_opt (sbi->s_mount_opt, OLDALLOC);
break;
#ifdef CONFIG_EXT4DEV_FS_XATTR
/* 配置文件擴展屬性 */
case Opt_user_xattr:
set_opt (sbi->s_mount_opt, XATTR_USER);
break;
/* 不配置文件擴展屬性 */
case Opt_nouser_xattr:
clear_opt (sbi->s_mount_opt, XATTR_USER);
break;
#else
case Opt_user_xattr:
case Opt_nouser_xattr:
printk("EXT4 (no)user_xattr options not supported\n");
break;
#endif
/* 文件控制權限列表的配置與否 */
#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
case Opt_acl:
set_opt(sbi->s_mount_opt, POSIX_ACL);
break;
case Opt_noacl:
clear_opt(sbi->s_mount_opt, POSIX_ACL);
break;
#else
case Opt_acl:
case Opt_noacl:
printk("EXT4 (no)acl options not supported\n");
break;
#endif
case Opt_reservation:
set_opt(sbi->s_mount_opt, RESERVATION);
break;
case Opt_noreservation:
clear_opt(sbi->s_mount_opt, RESERVATION);
break;
case Opt_journal_update:
/* 將ext4的日誌格式變成當前的格式 */
if (is_remount) {
printk(KERN_ERR "EXT4-fs: cannot specify "
"journal on remount\n");
return 0;
}
set_opt (sbi->s_mount_opt, UPDATE_JOURNAL);
break;
/* 如果是重新掛載不會理會這些journal選項，inum指定有多少個日誌文件 */
case Opt_journal_inum:
if (is_remount) {
printk(KERN_ERR "EXT4-fs: cannot specify "
"journal on remount\n");
return 0;
}
if (match_int(&args[0], &option))
return 0;
*inum = option;
break;
/* 指定日誌設備的主次設備號 */
case Opt_journal_dev:
if (is_remount) {
printk(KERN_ERR "EXT4-fs: cannot specify "
"journal on remount\n");
return 0;
}
if (match_int(&args[0], &option))
return 0;
*journal_devnum = option;
break;
/* 掛載的時候不會加載日誌 */
case Opt_noload:
set_opt (sbi->s_mount_opt, NOLOAD);
break;
/* 每過多少秒會把髒數據寫入到磁盤上 */
case Opt_commit:
if (match_int(&args[0], &option))
return 0;
if (option < 0)
return 0;
if (option == 0)
option = JBD_DEFAULT_MAX_COMMIT_AGE;
sbi->s_commit_interval = HZ * option;
break;
/* 三種數據模式的置爲和清除位 */
case Opt_data_journal:
data_opt = EXT4_MOUNT_JOURNAL_DATA;
goto datacheck;
case Opt_data_ordered:
data_opt = EXT4_MOUNT_ORDERED_DATA;
goto datacheck;
case Opt_data_writeback:
data_opt = EXT4_MOUNT_WRITEBACK_DATA;
datacheck:
/* 對數據位的檢查 */
if (is_remount) {
if ((sbi->s_mount_opt & EXT4_MOUNT_DATA_FLAGS)
!= data_opt) {
printk(KERN_ERR
"EXT4-fs: cannot change data "
"mode on remount\n");
return 0;
}
} else {
sbi->s_mount_opt &= ~EXT4_MOUNT_DATA_FLAGS;
sbi->s_mount_opt |= data_opt;
}
break;
#ifdef CONFIG_QUOTA
/* 用戶配額和組配額的應對函數 */
case Opt_usrjquota:
qtype = USRQUOTA;
goto set_qf_name;
case Opt_grpjquota:
qtype = GRPQUOTA;
set_qf_name:
/* 判斷是不是啓用了配額限制，如果已經啓用了，返回錯誤，現在應當尚未啓動 */
if (sb_any_quota_enabled(sb)) {
printk(KERN_ERR
"EXT4-fs: Cannot change journalled "
"quota options when quota turned on.\n");
return 0;
}
/* 配額限制的文件名字 */
qname = match_strdup(&args[0]);
if (!qname) {
printk(KERN_ERR
"EXT4-fs: not enough memory for "
"storing quotafile name.\n");
return 0;
}
if (sbi->s_qf_names[qtype] &&
strcmp(sbi->s_qf_names[qtype], qname)) {
printk(KERN_ERR
"EXT4-fs: %s quota file already "
"specified.\n", QTYPE2NAME(qtype));
kfree(qname);
return 0;
}
sbi->s_qf_names[qtype] = qname;
/* 檢查配額文件名是不是在根目錄下 */
if (strchr(sbi->s_qf_names[qtype], '/')) {
printk(KERN_ERR
"EXT4-fs: quotafile must be on "
"filesystem root.\n");
kfree(sbi->s_qf_names[qtype]);
sbi->s_qf_names[qtype] = NULL;
return 0;
}
/* 設置配額對應位已經啓動 */
set_opt(sbi->s_mount_opt, QUOTA);
break;
/* 去除配額限制 */
case Opt_offusrjquota:
qtype = USRQUOTA;
goto clear_qf_name;
case Opt_offgrpjquota:
qtype = GRPQUOTA;
clear_qf_name:
if (sb_any_quota_enabled(sb)) {
printk(KERN_ERR "EXT4-fs: Cannot change "
"journalled quota options when "
"quota turned on.\n");
return 0;
}
sbi->s_qf_names[qtype] = NULL;
break;
case Opt_jqfmt_vfsold:
sbi->s_jquota_fmt = QFMT_VFS_OLD;
break;
case Opt_jqfmt_vfsv0:
sbi->s_jquota_fmt = QFMT_VFS_V0;
break;
/* 用戶和組的配額限制對應的位操作，已經無配額限制的位操作 */
case Opt_quota:
case Opt_usrquota:
set_opt(sbi->s_mount_opt, QUOTA);
set_opt(sbi->s_mount_opt, USRQUOTA);
break;
case Opt_grpquota:
set_opt(sbi->s_mount_opt, QUOTA);
set_opt(sbi->s_mount_opt, GRPQUOTA);
break;
case Opt_noquota:
if (sb_any_quota_enabled(sb)) {
printk(KERN_ERR "EXT4-fs: Cannot change quota "
"options when quota turned on.\n");
return 0;
}
clear_opt(sbi->s_mount_opt, QUOTA);
clear_opt(sbi->s_mount_opt, USRQUOTA);
clear_opt(sbi->s_mount_opt, GRPQUOTA);
break;
#else
case Opt_quota:
case Opt_usrquota:
case Opt_grpquota:
case Opt_usrjquota:
case Opt_grpjquota:
case Opt_offusrjquota:
case Opt_offgrpjquota:
case Opt_jqfmt_vfsold:
case Opt_jqfmt_vfsv0:
printk(KERN_ERR
"EXT4-fs: journalled quota options not "
"supported.\n");
break;
case Opt_noquota:
break;
#endif
case Opt_abort:
set_opt(sbi->s_mount_opt, ABORT);
break;
case Opt_barrier:
if (match_int(&args[0], &option))
return 0;
if (option)
set_opt(sbi->s_mount_opt, BARRIER);
else
clear_opt(sbi->s_mount_opt, BARRIER);
break;
case Opt_ignore:
break;
/* 改變inode大小 */
case Opt_resize:
if (!is_remount) {
printk("EXT4-fs: resize option only available "
"for remount\n");
return 0;
}
if (match_int(&args[0], &option) != 0)
return 0;
*n_blocks_count = option;
break;
/* 緩衝區首部是不是啓用，一般都會啓用 */
case Opt_nobh:
set_opt(sbi->s_mount_opt, NOBH);
break;
case Opt_bh:
clear_opt(sbi->s_mount_opt, NOBH);
break;
/* extents文件格式 */
case Opt_extents:
set_opt (sbi->s_mount_opt, EXTENTS);
break;
default:
printk (KERN_ERR
"EXT4-fs: Unrecognized mount option \"%s\" "
"or missing value\n", p);
return 0;
}
}
#ifdef CONFIG_QUOTA
if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) {
if ((sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA) &&
sbi->s_qf_names[USRQUOTA])
clear_opt(sbi->s_mount_opt, USRQUOTA);
if ((sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA) &&
sbi->s_qf_names[GRPQUOTA])
clear_opt(sbi->s_mount_opt, GRPQUOTA);
/* 檢驗配額格式是不是正確 */
if ((sbi->s_qf_names[USRQUOTA] &&
(sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA)) ||
(sbi->s_qf_names[GRPQUOTA] &&
(sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA))) {
printk(KERN_ERR "EXT4-fs: old and new quota "
"format mixing.\n");
return 0;
}
if (!sbi->s_jquota_fmt) {
printk(KERN_ERR "EXT4-fs: journalled quota format "
"not specified.\n");
return 0;
}
} else {
/* 如果日誌配額限制沒有指定會打印警告 */
if (sbi->s_jquota_fmt) {
printk(KERN_ERR "EXT4-fs: journalled quota format "
"specified with no journalling "
"enabled.\n");
return 0;
}
}
#endif
return 1;
}
/* 填充ext4_super_block結構體裏的一些變量 */
static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
int read_only)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
int res = 0;
/* 首先檢查，如果有問題會打印錯誤 */
if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) {
/* 當前es版本大於最高版本，打印錯誤，重新以只讀格式掛載 */
printk (KERN_ERR "EXT4-fs warning: revision level too high, "
"forcing read-only mode\n");
res = MS_RDONLY;
}
/* 已經只讀掛載直接返回 */
if (read_only)
return res;
/* 不是合法的文件系統，提示需要運行e2fsck檢查修復 */
if (!(sbi->s_mount_state & EXT4_VALID_FS))
printk (KERN_WARNING "EXT4-fs warning: mounting unchecked fs, "
"running e2fsck is recommended\n");
else if ((sbi->s_mount_state & EXT4_ERROR_FS))
printk (KERN_WARNING
"EXT4-fs warning: mounting fs with errors, "
"running e2fsck is recommended\n");
/* 掛載次數超出最大限制 */
else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 &&
le16_to_cpu(es->s_mnt_count) >=
(unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
printk (KERN_WARNING
"EXT4-fs warning: maximal mount count reached, "
"running e2fsck is recommended\n");
else if (le32_to_cpu(es->s_checkinterval) &&
(le32_to_cpu(es->s_lastcheck) +
le32_to_cpu(es->s_checkinterval) <= get_seconds()))
printk (KERN_WARNING
"EXT4-fs warning: checktime reached, "
"running e2fsck is recommended\n");
/* 如果es結構體的s_max_mnt_count沒有設置，我們就設置爲默認的最大值 */
if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
/* 掛載數目加一，時間記錄，更新版本 */
es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1);
es->s_mtime = cpu_to_le32(get_seconds());
ext4_update_dynamic_rev(sb);
EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
/* 把對es結構體的修改寫入磁盤 */
ext4_commit_super(sb, es, 1);
/* 如果掛載選項裏有debug還需要打印調試信息 */
if (test_opt(sb, DEBUG))
printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%lu, "
"bpg=%lu, ipg=%lu, mo=%04lx]\n",
sb->s_blocksize,
sbi->s_groups_count,
EXT4_BLOCKS_PER_GROUP(sb),
EXT4_INODES_PER_GROUP(sb),
sbi->s_mount_opt);
printk(KERN_INFO "EXT4 FS on %s, ", sb->s_id);
if (EXT4_SB(sb)->s_journal->j_inode == NULL) {
char b[BDEVNAME_SIZE];
printk("external journal on %s\n",
bdevname(EXT4_SB(sb)->s_journal->j_dev, b));
} else {
printk("internal journal\n");
}
return res;
}
/* 掛載時候調用，調用之前super_block會被locked，用來檢查組描述符 */
static int ext4_check_descriptors (struct super_block * sb)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
ext4_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block);
ext4_fsblk_t last_block;
ext4_fsblk_t block_bitmap;
ext4_fsblk_t inode_bitmap;
ext4_fsblk_t inode_table;
struct ext4_group_desc * gdp = NULL;
int desc_block = 0;
int i;
ext4_debug ("Checking group descriptors");
/* 對每一個組進行檢查 */
for (i = 0; i < sbi->s_groups_count; i++)
{
/* 最後一組要進行塊對齊，其實一般情況所有組的大小都是一樣的 */
if (i == sbi->s_groups_count - 1)
last_block = ext4_blocks_count(sbi->s_es) - 1;
else
last_block = first_block +
(EXT4_BLOCKS_PER_GROUP(sb) - 1);
/* 如果組描述符是所在位置是一個塊的開始 */
if ((i % EXT4_DESC_PER_BLOCK(sb)) == 0)
gdp = (struct ext4_group_desc *)
sbi->s_group_desc[desc_block++]->b_data;
/* 逐個進行數據塊位圖，inode位圖，inode table的檢查 */
block_bitmap = ext4_block_bitmap(sb, gdp);
if (block_bitmap < first_block || block_bitmap > last_block)
{
ext4_error (sb, "ext4_check_descriptors",
"Block bitmap for group %d"
" not in group (block %llu)!",
i, block_bitmap);
return 0;
}
inode_bitmap = ext4_inode_bitmap(sb, gdp);
if (inode_bitmap < first_block || inode_bitmap > last_block)
{
ext4_error (sb, "ext4_check_descriptors",
"Inode bitmap for group %d"
" not in group (block %llu)!",
i, inode_bitmap);
return 0;
}
inode_table = ext4_inode_table(sb, gdp);
if (inode_table < first_block ||
inode_table + sbi->s_itb_per_group > last_block)
{
ext4_error (sb, "ext4_check_descriptors",
"Inode table for group %d"
" not in group (block %llu)!",
i, inode_table);
return 0;
}
first_block += EXT4_BLOCKS_PER_GROUP(sb);
gdp = (struct ext4_group_desc *)
((__u8 *)gdp + EXT4_DESC_SIZE(sb));
}
/* 更新free blocks和free inode的值 */
ext4_free_blocks_count_set(sbi->s_es, ext4_count_free_blocks(sb));
sbi->s_es->s_free_inodes_count=cpu_to_le32(ext4_count_free_inodes(sb));
return 1;
}
/* ext4_orphan_cleanup()遍歷在超級塊上的inode鏈表，尋找被刪除的inode.努力在recovery的時候刪除這些inode.
* 我們僅僅是對每一個inode執行iget()和iput()，如果我們碰巧指向了一個正在使用的inode或者是已經被刪除的inode都是很安全的，最壞的情況也不過是得到一條"bit already cleared”的消息來自ext4_free_inode()函數.
*唯一的我們可能指向一個錯誤的inode原因是e2fsck正在文件系統上運行，它一定已經執行過孤兒inode的清理工作了，所以我們就直接終止就可以了。
*/
static void ext4_orphan_cleanup (struct super_block * sb,
struct ext4_super_block * es)
{
unsigned int s_flags = sb->s_flags;
int nr_orphans = 0, nr_truncates = 0;
#ifdef CONFIG_QUOTA
int i;
#endif
/* 沒有孤兒inode需要清理 */
if (!es->s_last_orphan) {
jbd_debug(4, "no orphan inodes to clean up\n");
return;
}
/* 有錯誤，停止清理inode */
if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) {
if (es->s_last_orphan)
jbd_debug(1, "Errors on filesystem, "
"clearing orphan list.\n");
es->s_last_orphan = 0;
jbd_debug(1, "Skipping orphan recovery on fs with errors.\n");
return;
}
/* 只讀文件系統，首先變成可讀寫的，清理完成以後再變只讀的 */
if (s_flags & MS_RDONLY) {
printk(KERN_INFO "EXT4-fs: %s: orphan cleanup on readonly fs\n",
sb->s_id);
sb->s_flags &= ~MS_RDONLY;
}
#ifdef CONFIG_QUOTA
/* 打開配額限制 */
sb->s_flags |= MS_ACTIVE;
for (i = 0; i < MAXQUOTAS; i++) {
if (EXT4_SB(sb)->s_qf_names[i]) {
int ret = ext4_quota_on_mount(sb, i);
if (ret < 0)
printk(KERN_ERR
"EXT4-fs: Cannot turn on journalled "
"quota: error %d\n", ret);
}
}
#endif
/* 遍歷每一個inode進行清理工作 */
while (es->s_last_orphan) {
struct inode *inode;
/* 首先獲取inode */
if (!(inode =
ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) {
es->s_last_orphan = 0;
break;
}
list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan);
DQUOT_INIT(inode);
/* 記錄刪除的和截斷的數目 */
if (inode->i_nlink) {
printk(KERN_DEBUG
"%s: truncating inode %lu to %Ld bytes\n",
__FUNCTION__, inode->i_ino, inode->i_size);
jbd_debug(2, "truncating inode %lu to %Ld bytes\n",
inode->i_ino, inode->i_size);
ext4_truncate(inode);
nr_truncates++;
} else {
printk(KERN_DEBUG
"%s: deleting unreferenced inode %lu\n",
__FUNCTION__, inode->i_ino);
jbd_debug(2, "deleting unreferenced inode %lu\n",
inode->i_ino);
nr_orphans++;
}
/* 刪除inode如果引用計數爲0 */
iput(inode);
}
#define PLURAL(x) (x), ((x)==1) ? "" : "s"
/* 打印調試信息 */
if (nr_orphans)
printk(KERN_INFO "EXT4-fs: %s: %d orphan inode%s deleted\n",
sb->s_id, PLURAL(nr_orphans));
if (nr_truncates)
printk(KERN_INFO "EXT4-fs: %s: %d truncate%s cleaned up\n",
sb->s_id, PLURAL(nr_truncates));
#ifdef CONFIG_QUOTA
/* 關閉配額限制 */
for (i = 0; i < MAXQUOTAS; i++) {
if (sb_dqopt(sb)->files[i])
vfs_quota_off(sb, i);
}
#endif
/* 重新加載只讀狀態 */
sb->s_flags = s_flags;
}
#define log2(n) ffz(~(n))
/*
* 最大文件大小. 這是一個直接索引塊，簡潔索引塊，雙重間接索引塊的限制.
*/
static loff_t ext4_max_size(int bits)
{
loff_t res = EXT4_NDIR_BLOCKS;
const loff_t upper_limit = 0x1ff7fffd000LL;
res += 1LL << (bits-2);
res += 1LL << (2*(bits-2));
res += 1LL << (3*(bits-2));
res <<= bits;
if (res > upper_limit)
res = upper_limit;
return res;
}
/* 返回這個塊組描述符所在的塊的位置 */
static ext4_fsblk_t descriptor_loc(struct super_block *sb,
ext4_fsblk_t logical_sb_block, int nr)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
unsigned long bg, first_meta_bg;
int has_super = 0;
/* 第一個元數據塊組 */
first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg);
/* 如果ext4沒有元數據塊組，就直接返回超級塊所在的位置的後邊nr個 */
if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG) ||
nr < first_meta_bg)
return logical_sb_block + nr + 1;
/* 第bg哥塊組描述符 */
bg = sbi->s_desc_per_block * nr;
/* 有超級塊的話需要跳過第一個超級塊 */
if (ext4_bg_has_super(sb, bg))
has_super = 1;
return (has_super + ext4_group_first_block_no(sb, bg));
}
/* 本文件最重要的函數，填充super_block結構體以及ext4_sb_info結構體的各個字段，掛載過程中最重要的函數 */
static int ext4_fill_super (struct super_block *sb, void *data, int silent)
{
struct buffer_head * bh;
struct ext4_super_block *es = NULL;
struct ext4_sb_info *sbi;
ext4_fsblk_t block;
ext4_fsblk_t sb_block = get_sb_block(&data);
ext4_fsblk_t logical_sb_block;
unsigned long offset = 0;
unsigned int journal_inum = 0;
unsigned long journal_devnum = 0;
unsigned long def_mount_opts;
struct inode *root;
int blocksize;
int hblock;
int db_count;
int i;
int needs_recovery;
__le32 features;
__u64 blocks_count;
/* 爲ext4_sb_info分配空間，包括一些簡單的缺省值的初始賦值 */
sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
if (!sbi)
return -ENOMEM;
sb->s_fs_info = sbi;
sbi->s_mount_opt = 0;
sbi->s_resuid = EXT4_DEF_RESUID;
sbi->s_resgid = EXT4_DEF_RESGID;
unlock_kernel();
/* 最小的塊大小 */
blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE);
if (!blocksize) {
printk(KERN_ERR "EXT4-fs: unable to set blocksize\n");
goto out_fail;
}
/*
* ext4只有在1KB大小的塊情況下才會buffer對齊，如果不是1KB大小的塊我們需要計算在塊內的偏移值 */
if (blocksize != EXT4_MIN_BLOCK_SIZE) {
logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
offset = do_div(logical_sb_block, blocksize);
} else {
logical_sb_block = sb_block;
}
/* 讀取超級塊所在的塊 */
if (!(bh = sb_bread(sb, logical_sb_block))) {
printk (KERN_ERR "EXT4-fs: unable to read superblock\n");
goto out_fail;
}
/*
* s_es必須被儘快的初始化，因爲很多的ext4的宏會依賴這個結構體的值
*/
es = (struct ext4_super_block *) (((char *)bh->b_data) + offset);
sbi->s_es = es;
/* 超級塊的魔數，檢查是否合法 */
sb->s_magic = le16_to_cpu(es->s_magic);
if (sb->s_magic != EXT4_SUPER_MAGIC)
goto cantfind_ext4;
/* 在轉換掛載選項之前，先設置初始的值 */
def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
if (def_mount_opts & EXT4_DEFM_DEBUG)
set_opt(sbi->s_mount_opt, DEBUG);
if (def_mount_opts & EXT4_DEFM_BSDGROUPS)
set_opt(sbi->s_mount_opt, GRPID);
if (def_mount_opts & EXT4_DEFM_UID16)
set_opt(sbi->s_mount_opt, NO_UID32);
#ifdef CONFIG_EXT4DEV_FS_XATTR
if (def_mount_opts & EXT4_DEFM_XATTR_USER)
set_opt(sbi->s_mount_opt, XATTR_USER);
#endif
#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
if (def_mount_opts & EXT4_DEFM_ACL)
set_opt(sbi->s_mount_opt, POSIX_ACL);
#endif
if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA)
sbi->s_mount_opt |= EXT4_MOUNT_JOURNAL_DATA;
else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED)
sbi->s_mount_opt |= EXT4_MOUNT_ORDERED_DATA;
else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK)
sbi->s_mount_opt |= EXT4_MOUNT_WRITEBACK_DATA;
if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC)
set_opt(sbi->s_mount_opt, ERRORS_PANIC);
else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_RO)
set_opt(sbi->s_mount_opt, ERRORS_RO);
else
set_opt(sbi->s_mount_opt, ERRORS_CONT);
sbi->s_resuid = le16_to_cpu(es->s_def_resuid);
sbi->s_resgid = le16_to_cpu(es->s_def_resgid);
set_opt(sbi->s_mount_opt, RESERVATION);
/* 缺省掛載選項設置完畢以後，根據用戶傳入的掛載選項來設置 */
if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum,
NULL, 0))
goto failed_mount;
/* 超級塊結構體的flags和權限控制列表 */
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
(EXT4_HAS_COMPAT_FEATURE(sb, ~0U) ||
EXT4_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
EXT4_HAS_INCOMPAT_FEATURE(sb, ~0U)))
printk(KERN_WARNING
"EXT4-fs warning: feature flags set on rev 0 fs, "
"running e2fsck is recommended\n");
/*
* 檢查feature flags有沒有不支持的flag.
*/
features = EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT4_FEATURE_INCOMPAT_SUPP);
if (features) {
printk(KERN_ERR "EXT4-fs: %s: couldn't mount because of "
"unsupported optional features (%x).\n",
sb->s_id, le32_to_cpu(features));
goto failed_mount;
}
features = EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT4_FEATURE_RO_COMPAT_SUPP);
if (!(sb->s_flags & MS_RDONLY) && features) {
printk(KERN_ERR "EXT4-fs: %s: couldn't mount RDWR because of "
"unsupported optional features (%x).\n",
sb->s_id, le32_to_cpu(features));
goto failed_mount;
}
/* blocksize設置成爲用戶傳入的blocksize */
blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
/* 檢查blocksize是不是合法 */
if (blocksize < EXT4_MIN_BLOCK_SIZE ||
blocksize > EXT4_MAX_BLOCK_SIZE) {
printk(KERN_ERR
"EXT4-fs: Unsupported filesystem blocksize %d on %s.\n",
blocksize, sb->s_id);
goto failed_mount;
}
/* 得到塊設備上的塊大小 */
hblock = bdev_hardsect_size(sb->s_bdev);
/* 塊設備的塊大小和文件系統的塊大小不一致 */
if (sb->s_blocksize != blocksize) {
/*
* 確保文件系統塊大小大於硬件設備上的塊大小.
*/
if (blocksize < hblock) {
printk(KERN_ERR "EXT4-fs: blocksize %d too small for "
"device blocksize %d.\n", blocksize, hblock);
goto failed_mount;
}
brelse (bh);
/* 根據新的塊大小，得到新的超級塊和es結構體 */
sb_set_blocksize(sb, blocksize);
logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
offset = do_div(logical_sb_block, blocksize);
bh = sb_bread(sb, logical_sb_block);
if (!bh) {
printk(KERN_ERR
"EXT4-fs: Can't read superblock on 2nd try.\n");
goto failed_mount;
}
es = (struct ext4_super_block *)(((char *)bh->b_data) + offset);
sbi->s_es = es;
if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) {
printk (KERN_ERR
"EXT4-fs: Magic mismatch, very weird !\n");
goto failed_mount;
}
}
/* 文件系統的最大文件大小，調用了前邊的函數 */
sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits);
/* inode版本如果是老版本，inode和first ino都是老的值 */
if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) {
sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE;
sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO;
} else {
/* 指定的inode大小的偏移值 */
sbi->s_inode_size = le16_to_cpu(es->s_inode_size);
sbi->s_first_ino = le32_to_cpu(es->s_first_ino);
if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
(sbi->s_inode_size & (sbi->s_inode_size - 1)) ||
(sbi->s_inode_size > blocksize)) {
printk (KERN_ERR
"EXT4-fs: unsupported inode size: %d\n",
sbi->s_inode_size);
goto failed_mount;
}
}
/* 文件碎片的大小設置和檢查 */
sbi->s_frag_size = EXT4_MIN_FRAG_SIZE <<
le32_to_cpu(es->s_log_frag_size);
if (blocksize != sbi->s_frag_size) {
printk(KERN_ERR
"EXT4-fs: fragsize %lu != blocksize %u (unsupported)\n",
sbi->s_frag_size, blocksize);
goto failed_mount;
}
/* 組描述符大小 */
sbi->s_desc_size = le16_to_cpu(es->s_desc_size);
if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) {
if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT ||
sbi->s_desc_size > EXT4_MAX_DESC_SIZE ||
sbi->s_desc_size & (sbi->s_desc_size - 1)) {
printk(KERN_ERR
"EXT4-fs: unsupported descriptor size %lu\n",
sbi->s_desc_size);
goto failed_mount;
}
} else
sbi->s_desc_size = EXT4_MIN_DESC_SIZE;
/* 每組的塊，inode，碎片數目等字段賦值 */
sbi->s_blocks_per_group = le32_to_cpu(es->s_blocks_per_group);
sbi->s_frags_per_group = le32_to_cpu(es->s_frags_per_group);
sbi->s_inodes_per_group = le32_to_cpu(es->s_inodes_per_group);
if (EXT4_INODE_SIZE(sb) == 0)
goto cantfind_ext4;
sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb);
if (sbi->s_inodes_per_block == 0)
goto cantfind_ext4;
sbi->s_itb_per_group = sbi->s_inodes_per_group /
sbi->s_inodes_per_block;
sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb);
sbi->s_sbh = bh;
sbi->s_mount_state = le16_to_cpu(es->s_state);
sbi->s_addr_per_block_bits = log2(EXT4_ADDR_PER_BLOCK(sb));
sbi->s_desc_per_block_bits = log2(EXT4_DESC_PER_BLOCK(sb));
for (i=0; i < 4; i++)
sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
sbi->s_def_hash_version = es->s_def_hash_version;
/* 賦值過後的檢查，檢查結果非法就會跳轉到failed_mount */
if (sbi->s_blocks_per_group > blocksize * 8) {
printk (KERN_ERR
"EXT4-fs: #blocks per group too big: %lu\n",
sbi->s_blocks_per_group);
goto failed_mount;
}
if (sbi->s_frags_per_group > blocksize * 8) {
printk (KERN_ERR
"EXT4-fs: #fragments per group too big: %lu\n",
sbi->s_frags_per_group);
goto failed_mount;
}
if (sbi->s_inodes_per_group > blocksize * 8) {
printk (KERN_ERR
"EXT4-fs: #inodes per group too big: %lu\n",
sbi->s_inodes_per_group);
goto failed_mount;
}
if (ext4_blocks_count(es) >
(sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) {
printk(KERN_ERR "EXT4-fs: filesystem on %s:"
" too large to mount safely\n", sb->s_id);
if (sizeof(sector_t) < 8)
printk(KERN_WARNING "EXT4-fs: CONFIG_LBD not "
"enabled\n");
goto failed_mount;
}
if (EXT4_BLOCKS_PER_GROUP(sb) == 0)
goto cantfind_ext4;
/* 組的數目 */
blocks_count = (ext4_blocks_count(es) -
le32_to_cpu(es->s_first_data_block) +
EXT4_BLOCKS_PER_GROUP(sb) - 1);
do_div(blocks_count, EXT4_BLOCKS_PER_GROUP(sb));
sbi->s_groups_count = blocks_count;
/* 有多少塊存放塊組描述符 */
db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
EXT4_DESC_PER_BLOCK(sb);
/* 賦值sbi的塊組描述符結構體數組 */
sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *),
GFP_KERNEL);
if (sbi->s_group_desc == NULL) {
printk (KERN_ERR "EXT4-fs: not enough memory\n");
goto failed_mount;
}
/* 塊組描述符數組的鎖 */
bgl_lock_init(&sbi->s_blockgroup_lock);
/* 逐個對塊組描述符賦值 */
for (i = 0; i < db_count; i++) {
block = descriptor_loc(sb, logical_sb_block, i);
sbi->s_group_desc[i] = sb_bread(sb, block);
if (!sbi->s_group_desc[i]) {
printk (KERN_ERR "EXT4-fs: "
"can't read group descriptor %d\n", i);
db_count = i;
goto failed_mount2;
}
}
/* 檢查 */
if (!ext4_check_descriptors (sb)) {
printk(KERN_ERR "EXT4-fs: group descriptors corrupted!\n");
goto failed_mount2;
}
/* sbi的存放塊組描述符的塊數目和幾個單cpu計數器 */
sbi->s_gdb_count = db_count;
get_random_bytes(&sbi->s_next_generation, sizeof(u32));
spin_lock_init(&sbi->s_next_gen_lock);
percpu_counter_init(&sbi->s_freeblocks_counter,
ext4_count_free_blocks(sb));
percpu_counter_init(&sbi->s_freeinodes_counter,
ext4_count_free_inodes(sb));
percpu_counter_init(&sbi->s_dirs_counter,
ext4_count_dirs(sb));
/* 每個文件系統的保留快列表和鎖初始化 */
spin_lock_init(&sbi->s_rsv_window_lock);
sbi->s_rsv_window_root = RB_ROOT;
/* 保留塊窗口變量初始化. */
sbi->s_rsv_window_head.rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
sbi->s_rsv_window_head.rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
sbi->s_rsv_window_head.rsv_alloc_hit = 0;
sbi->s_rsv_window_head.rsv_goal_size = 0;
ext4_rsv_window_add(sb, &sbi->s_rsv_window_head);
/*
* 初始化能夠讀取一個inode的相關資源
*/
sb->s_op = &ext4_sops;
sb->s_export_op = &ext4_export_ops;
sb->s_xattr = ext4_xattr_handlers;
#ifdef CONFIG_QUOTA
sb->s_qcop = &ext4_qctl_operations;
sb->dq_op = &ext4_quota_operations;
#endif
INIT_LIST_HEAD(&sbi->s_orphan);
sb->s_root = NULL;
/* 有沒有錯誤需要recovery的 */
needs_recovery = (es->s_last_orphan != 0 ||
EXT4_HAS_INCOMPAT_FEATURE(sb,
EXT4_FEATURE_INCOMPAT_RECOVER));
/*
* 第一個創建的inode是journal的inode
*/
if (!test_opt(sb, NOLOAD) &&
EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) {
if (ext4_load_journal(sb, es, journal_devnum))
goto failed_mount3;
} else if (journal_inum) {
if (ext4_create_journal(sb, es, journal_inum))
goto failed_mount3;
} else {
if (!silent)
printk (KERN_ERR
"ext4: No journal on filesystem on %s\n",
sb->s_id);
goto failed_mount3;
}
/* 如果有要求的話，我們會更新日誌，然後檢查日誌模式. */
switch (test_opt(sb, DATA_FLAGS)) {
case 0:
/* 沒有日誌模式設置，根據日誌能力設置
*/
if (jbd2_journal_check_available_features
(sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE))
set_opt(sbi->s_mount_opt, ORDERED_DATA);
else
set_opt(sbi->s_mount_opt, JOURNAL_DATA);
break;
case EXT4_MOUNT_ORDERED_DATA:
case EXT4_MOUNT_WRITEBACK_DATA:
if (!jbd2_journal_check_available_features
(sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) {
printk(KERN_ERR "EXT4-fs: Journal does not support "
"requested data journaling mode\n");
goto failed_mount4;
}
default:
break;
}
/* 只有writeback模式下才能NOBH */
if (test_opt(sb, NOBH)) {
if (!(test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)) {
printk(KERN_WARNING "EXT4-fs: Ignoring nobh option - "
"its supported only with writeback mode\n");
clear_opt(sbi->s_mount_opt, NOBH);
}
}
/*
* 現在可以掛載文件系統了.分配根節點的inode和dentry
*/
root = iget(sb, EXT4_ROOT_INO);
sb->s_root = d_alloc_root(root);
if (!sb->s_root) {
printk(KERN_ERR "EXT4-fs: get root inode failed\n");
iput(root);
goto failed_mount4;
}
if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
dput(sb->s_root);
sb->s_root = NULL;
printk(KERN_ERR "EXT4-fs: corrupt root inode, run e2fsck\n");
goto failed_mount4;
}
ext4_setup_super (sb, es, sb->s_flags & MS_RDONLY);
EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS;
ext4_orphan_cleanup(sb, es);
EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS;
if (needs_recovery)
printk (KERN_INFO "EXT4-fs: recovery complete.\n");
ext4_mark_recovery_complete(sb, es);
printk (KERN_INFO "EXT4-fs: mounted filesystem with %s data mode.\n",
test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ? "journal":
test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered":
"writeback");
ext4_ext_init(sb);
lock_kernel();
return 0;
cantfind_ext4:
if (!silent)
printk(KERN_ERR "VFS: Can't find ext4 filesystem on dev %s.\n",
sb->s_id);
goto failed_mount;
failed_mount4:
jbd2_journal_destroy(sbi->s_journal);
failed_mount3:
percpu_counter_destroy(&sbi->s_freeblocks_counter);
percpu_counter_destroy(&sbi->s_freeinodes_counter);
percpu_counter_destroy(&sbi->s_dirs_counter);
failed_mount2:
for (i = 0; i < db_count; i++)
brelse(sbi->s_group_desc[i]);
kfree(sbi->s_group_desc);
failed_mount:
#ifdef CONFIG_QUOTA
for (i = 0; i < MAXQUOTAS; i++)
kfree(sbi->s_qf_names[i]);
#endif
ext4_blkdev_remove(sbi);
brelse(bh);
out_fail:
sb->s_fs_info = NULL;
kfree(sbi);
lock_kernel();
return -EINVAL;
}
/*
* 初始化日誌參數，這個操作不僅僅會在掛載的時候執行，在recovery和remount的時候也會執行.
*/
static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
/* 如果用戶傳入的commit參數是0的話，就設置爲默認的commit時間 */
if (sbi->s_commit_interval)
journal->j_commit_interval = sbi->s_commit_interval;
spin_lock(&journal->j_state_lock);
if (test_opt(sb, BARRIER))
journal->j_flags |= JBD2_BARRIER;
else
journal->j_flags &= ~JBD2_BARRIER;
spin_unlock(&journal->j_state_lock);
}
/* 返回ext4的日誌結構體 */
static journal_t *ext4_get_journal(struct super_block *sb,
unsigned int journal_inum)
{
struct inode *journal_inode;
journal_t *journal;
/* 獲得journal的inode，然後檢驗是不是引用計數爲0，這意味着被刪除了 */
journal_inode = iget(sb, journal_inum);
if (!journal_inode) {
printk(KERN_ERR "EXT4-fs: no journal found.\n");
return NULL;
}
if (!journal_inode->i_nlink) {
make_bad_inode(journal_inode);
iput(journal_inode);
printk(KERN_ERR "EXT4-fs: journal inode is deleted.\n");
return NULL;
}
jbd_debug(2, "Journal inode found at %p: %Ld bytes\n",
journal_inode, journal_inode->i_size);
/* 檢查inode是否合法 */
if (is_bad_inode(journal_inode) || !S_ISREG(journal_inode->i_mode)) {
printk(KERN_ERR "EXT4-fs: invalid journal inode.\n");
iput(journal_inode);
return NULL;
}
/* 調用jbd2模塊的初始化函數 */
journal = jbd2_journal_init_inode(journal_inode);
if (!journal) {
printk(KERN_ERR "EXT4-fs: Could not load journal inode\n");
iput(journal_inode);
return NULL;
}
journal->j_private = sb;
ext4_init_journal_params(sb, journal);
return journal;
}
/* 根據用戶傳入的主次設備號來獲得日誌結構體 */
static journal_t *ext4_get_dev_journal(struct super_block *sb,
dev_t j_dev)
{
struct buffer_head * bh;
journal_t *journal;
ext4_fsblk_t start;
ext4_fsblk_t len;
int hblock, blocksize;
ext4_fsblk_t sb_block;
unsigned long offset;
struct ext4_super_block * es;
struct block_device *bdev;
/* 根據主次設備號，獲得塊設備結構體 */
bdev = ext4_blkdev_get(j_dev);
if (bdev == NULL)
return NULL;
if (bd_claim(bdev, sb)) {
printk(KERN_ERR
"EXT4: failed to claim external journal device.\n");
blkdev_put(bdev);
return NULL;
}
/* 檢查ext4和日誌塊設備的塊大小是不是匹配，要求ext4塊大小一定要大於日誌塊設備的塊大小 */
blocksize = sb->s_blocksize;
hblock = bdev_hardsect_size(bdev);
if (blocksize < hblock) {
printk(KERN_ERR
"EXT4-fs: blocksize too small for journal device.\n");
goto out_bdev;
}
/* 首先從日誌塊設備上獲得超級塊 */
sb_block = EXT4_MIN_BLOCK_SIZE / blocksize;
offset = EXT4_MIN_BLOCK_SIZE % blocksize;
set_blocksize(bdev, blocksize);
if (!(bh = __bread(bdev, sb_block, blocksize))) {
printk(KERN_ERR "EXT4-fs: couldn't read superblock of "
"external journal\n");
goto out_bdev;
}
/* 然後獲得es結構體 */
es = (struct ext4_super_block *) (((char *)bh->b_data) + offset);
if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) ||
!(le32_to_cpu(es->s_feature_incompat) &
EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) {
printk(KERN_ERR "EXT4-fs: external journal has "
"bad superblock\n");
brelse(bh);
goto out_bdev;
}
/* 檢驗日誌的uuid是不是符合 */
if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
printk(KERN_ERR "EXT4-fs: journal UUID does not match\n");
brelse(bh);
goto out_bdev;
}
len = ext4_blocks_count(es);
start = sb_block + 1;
brelse(bh); /* 超級塊使用完畢 */
/* jbd2模塊的初始化操作 */
journal = jbd2_journal_init_dev(bdev, sb->s_bdev,
start, len, blocksize);
if (!journal) {
printk(KERN_ERR "EXT4-fs: failed to create device journal\n");
goto out_bdev;
}
journal->j_private = sb;
/* 經典函數，讀j_sb_buffer塊 */
ll_rw_block(READ, 1, &journal->j_sb_buffer);
/* 等待讀取完畢 */
wait_on_buffer(journal->j_sb_buffer);
/* 檢查是不是uptodate以及是不是僅僅一個user */
if (!buffer_uptodate(journal->j_sb_buffer)) {
printk(KERN_ERR "EXT4-fs: I/O error on journal device\n");
goto out_journal;
}
if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
printk(KERN_ERR "EXT4-fs: External journal has more than one "
"user (unsupported) - %d\n",
be32_to_cpu(journal->j_superblock->s_nr_users));
goto out_journal;
}
EXT4_SB(sb)->journal_bdev = bdev;
ext4_init_journal_params(sb, journal);
return journal;
out_journal:
jbd2_journal_destroy(journal);
out_bdev:
ext4_blkdev_put(bdev);
return NULL;
}
/* 根據日誌的主次設備號來進行加載日誌操作 */
static int ext4_load_journal(struct super_block *sb,
struct ext4_super_block *es,
unsigned long journal_devnum)
{
journal_t *journal;
unsigned int journal_inum = le32_to_cpu(es->s_journal_inum);
dev_t journal_dev;
int err = 0;
int really_read_only;
/* 如果指定了日誌設備就用指定的日誌設備 */
if (journal_devnum &&
journal_devnum != le32_to_cpu(es->s_journal_dev)) {
printk(KERN_INFO "EXT4-fs: external journal device major/minor "
"numbers have changed\n");
journal_dev = new_decode_dev(journal_devnum);
} else
/* 使用es的默認日誌塊設備 */
journal_dev = new_decode_dev(le32_to_cpu(es->s_journal_dev));
really_read_only = bdev_read_only(sb->s_bdev);
/*
* 如果我們在recovery，我們需要檢查是不是擁有對設備的讀寫權限.
*/
if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) {
if (sb->s_flags & MS_RDONLY) {
printk(KERN_INFO "EXT4-fs: INFO: recovery "
"required on readonly filesystem.\n");
if (really_read_only) {
printk(KERN_ERR "EXT4-fs: write access "
"unavailable, cannot proceed.\n");
return -EROFS;
}
printk (KERN_INFO "EXT4-fs: write access will "
"be enabled during recovery.\n");
}
}
/* journalinum和日誌塊設備不能共存 */
if (journal_inum && journal_dev) {
printk(KERN_ERR "EXT4-fs: filesystem has both journal "
"and inode journals!\n");
return -EINVAL;
}
/* 獲得日誌結構體 */
if (journal_inum) {
if (!(journal = ext4_get_journal(sb, journal_inum)))
return -EINVAL;
} else {
if (!(journal = ext4_get_dev_journal(sb, journal_dev)))
return -EINVAL;
}
/* 獲得日誌結構體以後，檢查是不是把日誌需要更新爲當前的格式 */
if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) {
err = jbd2_journal_update_format(journal);
if (err) {
printk(KERN_ERR "EXT4-fs: error updating journal.\n");
jbd2_journal_destroy(journal);
return err;
}
}
if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER))
err = jbd2_journal_wipe(journal, !really_read_only);
if (!err)
err = jbd2_journal_load(journal);
if (err) {
printk(KERN_ERR "EXT4-fs: error loading journal.\n");
jbd2_journal_destroy(journal);
return err;
}
EXT4_SB(sb)->s_journal = journal;
ext4_clear_journal_err(sb, es);
if (journal_devnum &&
journal_devnum != le32_to_cpu(es->s_journal_dev)) {
es->s_journal_dev = cpu_to_le32(journal_devnum);
sb->s_dirt = 1;
/* 把recovery的flag刷新到磁盤 */
ext4_commit_super(sb, es, 1);
}
return 0;
}
/* 創建日誌結構體 */
static int ext4_create_journal(struct super_block * sb,
struct ext4_super_block * es,
unsigned int journal_inum)
{
journal_t *journal;
/* 檢查，只讀情況下返回錯誤 */
if (sb->s_flags & MS_RDONLY) {
printk(KERN_ERR "EXT4-fs: readonly filesystem when trying to "
"create journal.\n");
return -EROFS;
}
/* 通過日誌的inode號碼返回日誌結構體 */
if (!(journal = ext4_get_journal(sb, journal_inum)))
return -EINVAL;
printk(KERN_INFO "EXT4-fs: creating new journal on inode %u\n",
journal_inum);
/* jbd2的真正創建的函數 */
if (jbd2_journal_create(journal)) {
printk(KERN_ERR "EXT4-fs: error creating journal.\n");
jbd2_journal_destroy(journal);
return -EIO;
}
/* 存儲日誌結構體 */
EXT4_SB(sb)->s_journal = journal;
ext4_update_dynamic_rev(sb);
/* feature的設置 */
EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
EXT4_SET_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL);
es->s_journal_inum = cpu_to_le32(journal_inum);
sb->s_dirt = 1;
/* 把recovery的flag寫到磁盤. */
ext4_commit_super(sb, es, 1);
return 0;
}
/* 把超級塊寫入到磁盤 */
static void ext4_commit_super (struct super_block * sb,
struct ext4_super_block * es,
int sync)
{
struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
if (!sbh)
return;
/* 寫入時間的修改 */
es->s_wtime = cpu_to_le32(get_seconds());
ext4_free_blocks_count_set(es, ext4_count_free_blocks(sb));
es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb));
BUFFER_TRACE(sbh, "marking dirty");
/* 標記緩衝區首部爲髒 */
mark_buffer_dirty(sbh);
/* 然後同步 */
if (sync)
sync_dirty_buffer(sbh);
}
/*
* recovry結束了以後，如果我們在mount或者remount爲只讀，我們就需要在日誌上記錄 */
static void ext4_mark_recovery_complete(struct super_block * sb,
struct ext4_super_block * es)
{
journal_t *journal = EXT4_SB(sb)->s_journal;
jbd2_journal_lock_updates(journal);
jbd2_journal_flush(journal);
if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) &&
sb->s_flags & MS_RDONLY) {
/* 正在recovery並且還是隻讀，把super的修改寫入到磁盤 */
EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
sb->s_dirt = 0;
ext4_commit_super(sb, es, 1);
}
jbd2_journal_unlock_updates(journal);
}
/*
* 如果我們掛載的fs記錄了一個錯誤，我們需要把錯誤放到主文件系統中
*/
static void ext4_clear_journal_err(struct super_block * sb,
struct ext4_super_block * es)
{
journal_t *journal;
int j_errno;
const char *errstr;
journal = EXT4_SB(sb)->s_journal;
/*
* 檢查狀態，如果有錯誤，用ext4_error()或者ext4_abort()
*/
j_errno = jbd2_journal_errno(journal);
if (j_errno) {
char nbuf[16];
/* 把錯誤轉換成字符串 */
errstr = ext4_decode_error(sb, j_errno, nbuf);
/* 打印錯誤 */
ext4_warning(sb, __FUNCTION__, "Filesystem error recorded "
"from previous mount: %s", errstr);
ext4_warning(sb, __FUNCTION__, "Marking fs in need of "
"filesystem check.");
/* es和sbi結構體都要記錄錯誤，然後把超級塊寫入到磁盤 */
EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
ext4_commit_super (sb, es, 1);
/* 記錄結束以後，清理錯誤字段 */
jbd2_journal_clear_err(journal);
}
}
/*
* 強制提交事務，等待結束.
*/
int ext4_force_commit(struct super_block *sb)
{
journal_t *journal;
int ret;
if (sb->s_flags & MS_RDONLY)
return 0;
journal = EXT4_SB(sb)->s_journal;
sb->s_dirt = 0;
ret = ext4_journal_force_commit(journal);
return ret;
}
/*
* Ext4通過日誌寫入超級塊,我們不必手動寫入，僅僅是啓動writeback即可
*/
static void ext4_write_super (struct super_block * sb)
{
if (mutex_trylock(&sb->s_lock) != 0)
BUG();
sb->s_dirt = 0;
}
static int ext4_sync_fs(struct super_block *sb, int wait)
{
tid_t target;
sb->s_dirt = 0;
if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) {
if (wait)
jbd2_log_wait_commit(EXT4_SB(sb)->s_journal, target);
}
return 0;
}
/*
* LVM會調用這個函數，在一個只讀快照創建之前。這個函數會把日誌全部寫入磁盤.
*/
static void ext4_write_super_lockfs(struct super_block *sb)
{
sb->s_dirt = 0;
if (!(sb->s_flags & MS_RDONLY)) {
journal_t *journal = EXT4_SB(sb)->s_journal;
/* 啓動日誌barriar. */
jbd2_journal_lock_updates(journal);
jbd2_journal_flush(journal);
EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
}
}
/*
* 當LVM的日誌創建操作結束以後調用，重新設置RECOVER的flag
*/
static void ext4_unlockfs(struct super_block *sb)
{
if (!(sb->s_flags & MS_RDONLY)) {
lock_super(sb);
EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1);
unlock_super(sb);
jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
}
}
/* remount的函數 */
static int ext4_remount (struct super_block * sb, int * flags, char * data)
{
struct ext4_super_block * es;
struct ext4_sb_info *sbi = EXT4_SB(sb);
ext4_fsblk_t n_blocks_count = 0;
unsigned long old_sb_flags;
struct ext4_mount_options old_opts;
int err;
#ifdef CONFIG_QUOTA
int i;
#endif
/* 存儲原先的掛載選項 */
old_sb_flags = sb->s_flags;
old_opts.s_mount_opt = sbi->s_mount_opt;
old_opts.s_resuid = sbi->s_resuid;
old_opts.s_resgid = sbi->s_resgid;
old_opts.s_commit_interval = sbi->s_commit_interval;
#ifdef CONFIG_QUOTA
old_opts.s_jquota_fmt = sbi->s_jquota_fmt;
for (i = 0; i < MAXQUOTAS; i++)
old_opts.s_qf_names[i] = sbi->s_qf_names[i];
#endif
/*
* 檢查選項從而使用新的掛載選項.
*/
if (!parse_options(data, sb, NULL, NULL, &n_blocks_count, 1)) {
err = -EINVAL;
goto restore_opts;
}
if (sbi->s_mount_opt & EXT4_MOUNT_ABORT)
ext4_abort(sb, __FUNCTION__, "Abort forced by user");
sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
es = sbi->s_es;
/* 初始化新的日誌參數 */
ext4_init_journal_params(sb, sbi->s_journal);
if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY) ||
n_blocks_count > ext4_blocks_count(es)) {
if (sbi->s_mount_opt & EXT4_MOUNT_ABORT) {
err = -EROFS;
goto restore_opts;
}
/* 要只讀掛載 */
if (*flags & MS_RDONLY) {
/*
* 首先設置超級塊的flag爲只讀的
*/
sb->s_flags |= MS_RDONLY;
if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) &&
(sbi->s_mount_state & EXT4_VALID_FS))
es->s_state = cpu_to_le16(sbi->s_mount_state);
ext4_mark_recovery_complete(sb, es);
} else {
__le32 ret;
/* 如果原先是隻讀的，並且不支持非只讀掛載，打印錯誤 */
if ((ret = EXT4_HAS_RO_COMPAT_FEATURE(sb,
~EXT4_FEATURE_RO_COMPAT_SUPP))) {
printk(KERN_WARNING "EXT4-fs: %s: couldn't "
"remount RDWR because of unsupported "
"optional features (%x).\n",
sb->s_id, le32_to_cpu(ret));
err = -EROFS;
goto restore_opts;
}
/*
* 掛載文件系統爲可讀寫的
*/
ext4_clear_journal_err(sb, es);
sbi->s_mount_state = le16_to_cpu(es->s_state);
if ((err = ext4_group_extend(sb, es, n_blocks_count)))
goto restore_opts;
if (!ext4_setup_super (sb, es, 0))
sb->s_flags &= ~MS_RDONLY;
}
}
#ifdef CONFIG_QUOTA
/* 釋放原先的配額限制字符串 */
for (i = 0; i < MAXQUOTAS; i++)
if (old_opts.s_qf_names[i] &&
old_opts.s_qf_names[i] != sbi->s_qf_names[i])
kfree(old_opts.s_qf_names[i]);
#endif
return 0;
restore_opts:
sb->s_flags = old_sb_flags;
sbi->s_mount_opt = old_opts.s_mount_opt;
sbi->s_resuid = old_opts.s_resuid;
sbi->s_resgid = old_opts.s_resgid;
sbi->s_commit_interval = old_opts.s_commit_interval;
#ifdef CONFIG_QUOTA
/* 新的配額字符串 */
sbi->s_jquota_fmt = old_opts.s_jquota_fmt;
for (i = 0; i < MAXQUOTAS; i++) {
if (sbi->s_qf_names[i] &&
old_opts.s_qf_names[i] != sbi->s_qf_names[i])
kfree(sbi->s_qf_names[i]);
sbi->s_qf_names[i] = old_opts.s_qf_names[i];
}
#endif
return err;
}
/* 返回ext4的文件系統狀態，存放在buf裏 */
static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf)
{
struct super_block *sb = dentry->d_sb;
struct ext4_sb_info *sbi = EXT4_SB(sb);
struct ext4_super_block *es = sbi->s_es;
ext4_fsblk_t overhead;
int i;
/* 如果是minix的文件系統，開始就是狀態信息，不需要向後移動 */
if (test_opt (sb, MINIX_DF))
overhead = 0;
else {
unsigned long ngroups;
/* 組的數目 */
ngroups = EXT4_SB(sb)->s_groups_count;
smp_rmb();
/*
* 計算偏移
*/
/*
* first_data_block之前的都是數據偏移*/
overhead = le32_to_cpu(es->s_first_data_block);
/*
* Add the overhead attributed to the superblock and
* block group descriptors. If the sparse superblocks
* feature is turned on, then not all groups have this.
*/
for (i = 0; i < ngroups; i++) {
overhead += ext4_bg_has_super(sb, i) +
ext4_bg_num_gdb(sb, i);
cond_resched();
}
/*
* 每一個塊組都有inode位圖，數據塊位圖，inode table
*/
overhead += (ngroups * (2 + EXT4_SB(sb)->s_itb_per_group));
}
/* buf的賦值 */
buf->f_type = EXT4_SUPER_MAGIC;
buf->f_bsize = sb->s_blocksize;
buf->f_blocks = ext4_blocks_count(es) - overhead;
buf->f_bfree = percpu_counter_sum(&sbi->s_freeblocks_counter);
buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es);
if (buf->f_bfree < ext4_r_blocks_count(es))
buf->f_bavail = 0;
buf->f_files = le32_to_cpu(es->s_inodes_count);
buf->f_ffree = percpu_counter_sum(&sbi->s_freeinodes_counter);
buf->f_namelen = EXT4_NAME_LEN;
return 0;
}
/* 幫助配額同步的函數，在配額文件寫入之前我們必須鎖住 */
#ifdef CONFIG_QUOTA
/* 返回配額文件的inode */
static inline struct inode *dquot_to_inode(struct dquot *dquot)
{
return sb_dqopt(dquot->dq_sb)->files[dquot->dq_type];
}
/* 配額限制的初始化 */
static int ext4_dquot_initialize(struct inode *inode, int type)
{
handle_t *handle;
int ret, err;
/* 我們創建配額結構體，所以需要足夠的保留塊 */
handle = ext4_journal_start(inode, 2*EXT4_QUOTA_INIT_BLOCKS(inode->i_sb));
if (IS_ERR(handle))
return PTR_ERR(handle);
/* 配額初始化函數 */
ret = dquot_initialize(inode, type);
err = ext4_journal_stop(handle);
if (!ret)
ret = err;
return ret;
}
/* 釋放配額結構體 */
static int ext4_dquot_drop(struct inode *inode)
{
handle_t *handle;
int ret, err;
handle = ext4_journal_start(inode, 2*EXT4_QUOTA_DEL_BLOCKS(inode->i_sb));
if (IS_ERR(handle))
return PTR_ERR(handle);
ret = dquot_drop(inode);
err = ext4_journal_stop(handle);
if (!ret)
ret = err;
return ret;
}
/* 寫入配額結構體 */
static int ext4_write_dquot(struct dquot *dquot)
{
int ret, err;
handle_t *handle;
struct inode *inode;
/* 配額文件的inode */
inode = dquot_to_inode(dquot);
handle = ext4_journal_start(inode,
EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
if (IS_ERR(handle))
return PTR_ERR(handle);
/* 寫入函數 */
ret = dquot_commit(dquot);
err = ext4_journal_stop(handle);
if (!ret)
ret = err;
return ret;
}
/* 獲得配額結構體 */
static int ext4_acquire_dquot(struct dquot *dquot)
{
int ret, err;
handle_t *handle;
handle = ext4_journal_start(dquot_to_inode(dquot),
EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb));
if (IS_ERR(handle))
return PTR_ERR(handle);
ret = dquot_acquire(dquot);
err = ext4_journal_stop(handle);
if (!ret)
ret = err;
return ret;
}
/* 釋放配額結構體 */
static int ext4_release_dquot(struct dquot *dquot)
{
int ret, err;
handle_t *handle;
handle = ext4_journal_start(dquot_to_inode(dquot),
EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb));
if (IS_ERR(handle))
return PTR_ERR(handle);
ret = dquot_release(dquot);
err = ext4_journal_stop(handle);
if (!ret)
ret = err;
return ret;
}
/* 標記配額結構體髒 */
static int ext4_mark_dquot_dirty(struct dquot *dquot)
{
/* 判斷我們會不會對配額進行日誌操作 */
if (EXT4_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] ||
EXT4_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) {
dquot_mark_dquot_dirty(dquot);
return ext4_write_dquot(dquot);
} else {
return dquot_mark_dquot_dirty(dquot);
}
}
/* 配額信息寫入 */
static int ext4_write_info(struct super_block *sb, int type)
{
int ret, err;
handle_t *handle;
/* 數據塊+inode塊一共兩個 */
handle = ext4_journal_start(sb->s_root->d_inode, 2);
if (IS_ERR(handle))
return PTR_ERR(handle);
/* 調用dquot的函數提交信息 */
ret = dquot_commit_info(sb, type);
err = ext4_journal_stop(handle);
if (!ret)
ret = err;
return ret;
}
/*
* 打開配額
*/
static int ext4_quota_on_mount(struct super_block *sb, int type)
{
return vfs_quota_on_mount(sb, EXT4_SB(sb)->s_qf_names[type],
EXT4_SB(sb)->s_jquota_fmt, type);
}
/*
* 配額打開的時候調用的標準函數
*/
static int ext4_quota_on(struct super_block *sb, int type, int format_id,
char *path)
{
int err;
struct nameidata nd;
/* 不支持配額就直接退出 */
if (!test_opt(sb, QUOTA))
return -EINVAL;
/* 沒有找到配額限制 */
if (!EXT4_SB(sb)->s_qf_names[USRQUOTA] &&
!EXT4_SB(sb)->s_qf_names[GRPQUOTA])
return vfs_quota_on(sb, type, format_id, path);
err = path_lookup(path, LOOKUP_FOLLOW, &nd);
if (err)
return err;
/* 配額文件不在這個文件系統裏 */
if (nd.mnt->mnt_sb != sb) {
path_release(&nd);
return -EXDEV;
}
/* 配額文件不在根目錄 */
if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode)
printk(KERN_WARNING
"EXT4-fs: Quota file not on filesystem root. "
"Journalled quota will not work.\n");
path_release(&nd);
return vfs_quota_on(sb, type, format_id, path);
}
/* 從配額文件裏讀取配額限制，不從pagecache中讀取，這樣就避免了獲取鎖，因爲配額文件不會被截斷 */
static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data,
size_t len, loff_t off)
{
struct inode *inode = sb_dqopt(sb)->files[type];
/* 計算出來配額文件的塊數和塊內偏移 */
sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
int err = 0;
int offset = off & (sb->s_blocksize - 1);
int tocopy;
size_t toread;
struct buffer_head *bh;
loff_t i_size = i_size_read(inode);
if (off > i_size)
return 0;
if (off+len > i_size)
len = i_size-off;
toread = len;
/* 循環讀取配額文件內容 */
while (toread > 0) {
tocopy = sb->s_blocksize - offset < toread ?
sb->s_blocksize - offset : toread;
bh = ext4_bread(NULL, inode, blk, 0, &err);
if (err)
return err;
if (!bh) /* A hole? */
memset(data, 0, tocopy);
else
memcpy(data, bh->b_data+offset, tocopy);
brelse(bh);
offset = 0;
toread -= tocopy;
data += tocopy;
blk++;
}
return len;
}
/* 寫入配額文件 */
static ssize_t ext4_quota_write(struct super_block *sb, int type,
const char *data, size_t len, loff_t off)
{
struct inode *inode = sb_dqopt(sb)->files[type];
/* 計算出來配額文件的塊數和塊內偏移 */
sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
int err = 0;
int offset = off & (sb->s_blocksize - 1);
int tocopy;
int journal_quota = EXT4_SB(sb)->s_qf_names[type] != NULL;
size_t towrite = len;
struct buffer_head *bh;
handle_t *handle = journal_current_handle();
/* 寫入之前先鎖，然後循環讀取配額，然後修改內容 */
mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA);
while (towrite > 0) {
tocopy = sb->s_blocksize - offset < towrite ?
sb->s_blocksize - offset : towrite;
/* 讀取至緩衝區首部 */
bh = ext4_bread(handle, inode, blk, 1, &err);
if (!bh)
goto out;
if (journal_quota) {
err = ext4_journal_get_write_access(handle, bh);
if (err) {
brelse(bh);
goto out;
}
}
lock_buffer(bh);
/* 寫入修改 */
memcpy(bh->b_data+offset, data, tocopy);
flush_dcache_page(bh->b_page);
unlock_buffer(bh);
/* 如果有配額日誌的話就要日誌修改 */
if (journal_quota)
err = ext4_journal_dirty_metadata(handle, bh);
else {
/* Always do at least ordered writes for quotas */
err = ext4_journal_dirty_data(handle, bh);
mark_buffer_dirty(bh);
}
brelse(bh);
if (err)
goto out;
offset = 0;
towrite -= tocopy;
data += tocopy;
blk++;
}
out:
if (len == towrite)
return err;
if (inode->i_size < off+len-towrite) {
i_size_write(inode, off+len-towrite);
EXT4_I(inode)->i_disksize = inode->i_size;
}
inode->i_version++;
inode->i_mtime = inode->i_ctime = CURRENT_TIME;
ext4_mark_inode_dirty(handle, inode);
mutex_unlock(&inode->i_mutex);
return len - towrite;
}
#endif
/* 獲取超級塊 */
static int ext4_get_sb(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data, struct vfsmount *mnt)
{
return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt);
}
/* 註冊文件系統的結構體 */
static struct file_system_type ext4dev_fs_type = {
.owner = THIS_MODULE,
.name = "ext4dev",
.get_sb = ext4_get_sb,
.kill_sb = kill_block_super,
.fs_flags = FS_REQUIRES_DEV,
};
/* ext4的模塊初始化函數 */
static int __init init_ext4_fs(void)
{
/* 一次初始化擴展屬性，inode緩存 */
int err = init_ext4_xattr();
if (err)
return err;
err = init_inodecache();
if (err)
goto out1;
/* 註冊ext4文件系統 */
err = register_filesystem(&ext4dev_fs_type);
if (err)
goto out;
return 0;
out:
destroy_inodecache();
out1:
exit_ext4_xattr();
return err;
}
static void __exit exit_ext4_fs(void)
{
unregister_filesystem(&ext4dev_fs_type);
destroy_inodecache();
exit_ext4_xattr();
}
MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
MODULE_DESCRIPTION("Fourth Extended Filesystem with extents");
MODULE_LICENSE("GPL");
module_init(init_ext4_fs)
module_exit(exit_ext4_fs)