inode.c

基于组件方式开发操作系统的OSKIT源代码

/*
 * linux/fs/inode.c
 *
 * (C) 1997 Linus Torvalds
 *
 *  OSKit support added by the University of Utah, 1997
 */

#include <linux/fs.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/dcache.h>
#include <linux/init.h>
#include <linux/quotaops.h>

/*
 * New inode.c implementation.
 *
 * This implementation has the basic premise of trying
 * to be extremely low-overhead and SMP-safe, yet be
 * simple enough to be "obviously correct".
 *
 * Famous last words.
 */

#define INODE_PARANOIA 1
/* #define INODE_DEBUG 1 */

/*
 * Inode lookup is no longer as critical as it used to be:
 * most of the lookups are going to be through the dcache.
 */
#define HASH_BITS	8
#define HASH_SIZE	(1UL << HASH_BITS)
#define HASH_MASK	(HASH_SIZE-1)

/*
 * Each inode can be on two separate lists. One is
 * the hash list of the inode, used for lookups. The
 * other linked list is the "type" list:
 *  "in_use" - valid inode, hashed if i_nlink > 0
 *  "dirty"  - valid inode, hashed if i_nlink > 0, dirty.
 *  "unused" - ready to be re-used. Not hashed.
 *
 * A "dirty" list is maintained for each super block,
 * allowing for low-overhead inode sync() operations.
 */

LIST_HEAD(inode_in_use);
static LIST_HEAD(inode_unused);
static struct list_head inode_hashtable[HASH_SIZE];

/*
 * A simple spinlock to protect the list manipulations.
 *
 * NOTE! You also have to own the lock if you change
 * the i_state of an inode while it is in use..
 */
spinlock_t inode_lock = SPIN_LOCK_UNLOCKED;

/*
 * Statistics gathering..
 */
struct {
	int nr_inodes;
	int nr_free_inodes;
	int dummy[5];
} inodes_stat = {0, 0,};

int max_inodes;

/*
 * Put the inode on the super block's dirty list.
 *
 * CAREFUL! We mark it dirty unconditionally, but
 * move it onto the dirty list only if it is hashed.
 * If it was not hashed, it will never be added to
 * the dirty list even if it is later hashed, as it
 * will have been marked dirty already.
 *
 * In short, make sure you hash any inodes _before_
 * you start marking them dirty..
 */
void __mark_inode_dirty(struct inode *inode)
{
	struct super_block * sb = inode->i_sb;

	if (sb) {
		spin_lock(&inode_lock);
		if (!(inode->i_state & I_DIRTY)) {
			inode->i_state |= I_DIRTY;
			/* Only add valid (ie hashed) inodes to the dirty list */
			if (!list_empty(&inode->i_hash)) {
				list_del(&inode->i_list);
				list_add(&inode->i_list, &sb->s_dirty);
			}
		}
		spin_unlock(&inode_lock);
	}
}

static void __wait_on_inode(struct inode * inode)
{
	struct wait_queue wait = { current, NULL };

	add_wait_queue(&inode->i_wait, &wait);
repeat:
	current->state = TASK_UNINTERRUPTIBLE;
	if (inode->i_state & I_LOCK) {
		schedule();
		goto repeat;
	}
	remove_wait_queue(&inode->i_wait, &wait);
	current->state = TASK_RUNNING;
}

static inline void wait_on_inode(struct inode *inode)
{
	if (inode->i_state & I_LOCK)
		__wait_on_inode(inode);
}

/*
 * These are initializations that only need to be done
 * once, because the fields are idempotent across use
 * of the inode..
 */
static inline void init_once(struct inode * inode)
{
	memset(inode, 0, sizeof(*inode));
	init_waitqueue(&inode->i_wait);
	INIT_LIST_HEAD(&inode->i_hash);
	INIT_LIST_HEAD(&inode->i_dentry);
	sema_init(&inode->i_sem, 1);
	sema_init(&inode->i_atomic_write, 1);
}

static inline void write_inode(struct inode *inode)
{
	if (inode->i_sb && inode->i_sb->s_op && inode->i_sb->s_op->write_inode)
		inode->i_sb->s_op->write_inode(inode);
}

static inline void sync_one(struct inode *inode)
{
	if (inode->i_state & I_LOCK) {
		spin_unlock(&inode_lock);
		__wait_on_inode(inode);
		spin_lock(&inode_lock);
	} else {
		list_del(&inode->i_list);
		list_add(&inode->i_list, &inode_in_use);
		/* Set I_LOCK, reset I_DIRTY */
		inode->i_state ^= I_DIRTY | I_LOCK;
		spin_unlock(&inode_lock);

		write_inode(inode);

		spin_lock(&inode_lock);
		inode->i_state &= ~I_LOCK;
		wake_up(&inode->i_wait);
	}
}

static inline void sync_list(struct list_head *head)
{
	struct list_head * tmp;

	while ((tmp = head->prev) != head)
		sync_one(list_entry(tmp, struct inode, i_list));
}

/*
 * "sync_inodes()" goes through the super block's dirty list, 
 * writes them out, and puts them back on the normal list.
 */
void sync_inodes(kdev_t dev)
{
	struct super_block * sb = sb_entry(super_blocks.next);

	/*
	 * Search the super_blocks array for the device(s) to sync.
	 */
	spin_lock(&inode_lock);
	for (; sb != sb_entry(&super_blocks); sb = sb_entry(sb->s_list.next)) {
		if (!sb->s_dev)
			continue;
		if (dev && sb->s_dev != dev)
			continue;

		sync_list(&sb->s_dirty);

		if (dev)
			break;
	}
	spin_unlock(&inode_lock);
}

/*
 * Called with the spinlock already held..
 */
static void sync_all_inodes(void)
{
	struct super_block * sb = sb_entry(super_blocks.next);
	for (; sb != sb_entry(&super_blocks); sb = sb_entry(sb->s_list.next)) {
		if (!sb->s_dev)
			continue;
		sync_list(&sb->s_dirty);
	}
}

/*
 * Needed by knfsd
 */
void write_inode_now(struct inode *inode)
{
	struct super_block * sb = inode->i_sb;

	if (sb) {
		spin_lock(&inode_lock);
		while (inode->i_state & I_DIRTY)
			sync_one(inode);
		spin_unlock(&inode_lock);
	}
	else
		printk("write_inode_now: no super block\n");
}

/*
 * This is called by the filesystem to tell us
 * that the inode is no longer useful. We just
 * terminate it with extreme prejudice.
 */
void clear_inode(struct inode *inode)
{
	if (inode->i_nrpages)
		truncate_inode_pages(inode, 0);
	wait_on_inode(inode);
	if (IS_QUOTAINIT(inode))
		DQUOT_DROP(inode);
	if (inode->i_sb && inode->i_sb->s_op && inode->i_sb->s_op->clear_inode)
		inode->i_sb->s_op->clear_inode(inode);

	inode->i_state = 0;
}

/*
 * Dispose-list gets a local list, so it doesn't need to
 * worry about list corruption. It releases the inode lock
 * while clearing the inodes.
 */
static void dispose_list(struct list_head * head)
{
	struct list_head *next;
	int count = 0;

	spin_unlock(&inode_lock);
	next = head->next;
	for (;;) {
		struct list_head * tmp = next;
		struct inode * inode;

		next = next->next;
		if (tmp == head)
			break;
		inode = list_entry(tmp, struct inode, i_list);
		clear_inode(inode);
		count++;
	}

	/* Add them all to the unused list in one fell swoop */
	spin_lock(&inode_lock);
	list_splice(head, &inode_unused);
	inodes_stat.nr_free_inodes += count;
}

/*
 * Invalidate all inodes for a device.
 */
static int invalidate_list(struct list_head *head, struct super_block * sb, struct list_head * dispose)
{
	struct list_head *next;
	int busy = 0;

	next = head->next;
	for (;;) {
		struct list_head * tmp = next;
		struct inode * inode;

		next = next->next;
		if (tmp == head)
			break;
		inode = list_entry(tmp, struct inode, i_list);
		if (inode->i_sb != sb)
			continue;
		if (!inode->i_count) {
			list_del(&inode->i_hash);
			INIT_LIST_HEAD(&inode->i_hash);
			list_del(&inode->i_list);
			list_add(&inode->i_list, dispose);
			inode->i_state |= I_FREEING;
			continue;
		}
		busy = 1;
	}
	return busy;
}

/*
 * This is a two-stage process. First we collect all
 * offending inodes onto the throw-away list, and in
 * the second stage we actually dispose of them. This
 * is because we don't want to sleep while messing
 * with the global lists..
 */
int invalidate_inodes(struct super_block * sb)
{
	int busy;
	LIST_HEAD(throw_away);

	spin_lock(&inode_lock);
	busy = invalidate_list(&inode_in_use, sb, &throw_away);
	busy |= invalidate_list(&sb->s_dirty, sb, &throw_away);
	dispose_list(&throw_away);
	spin_unlock(&inode_lock);

	return busy;
}

/*
 * This is called with the inode lock held. It searches
 * the in-use for freeable inodes, which are moved to a
 * temporary list and then placed on the unused list by
 * dispose_list. 
 *
 * We don't expect to have to call this very often.
 *
 * N.B. The spinlock is released during the call to
 *      dispose_list.
 */
#define CAN_UNUSE(inode) \
	(((inode)->i_count | (inode)->i_state) == 0)
#define INODE(entry)	(list_entry(entry, struct inode, i_list))

static int free_inodes(void)
{
	struct list_head list, *entry, *freeable = &list;
	int found = 0;

	INIT_LIST_HEAD(freeable);
	entry = inode_in_use.next;
	while (entry != &inode_in_use) {
		struct list_head *tmp = entry;

		entry = entry->next;
		if (!CAN_UNUSE(INODE(tmp)))
			continue;
		list_del(tmp);
		list_del(&INODE(tmp)->i_hash);
		INIT_LIST_HEAD(&INODE(tmp)->i_hash);
		list_add(tmp, freeable);
		list_entry(tmp, struct inode, i_list)->i_state = I_FREEING;
		found = 1;
	}

	if (found)
		dispose_list(freeable);

	return found;
}

/*
 * Searches the inodes list for freeable inodes,
 * shrinking the dcache before (and possible after,
 * if we're low)
 */
static void try_to_free_inodes(int goal)
{
	/*
	 * First stry to just get rid of unused inodes.
	 *
	 * If we can't reach our goal that way, we'll have
	 * to try to shrink the dcache and sync existing
	 * inodes..
	 */
	free_inodes();
	goal -= inodes_stat.nr_free_inodes;
	if (goal > 0) {
		spin_unlock(&inode_lock);
		select_dcache(goal, 0);
		prune_dcache(goal);
		spin_lock(&inode_lock);
		sync_all_inodes();
		free_inodes();
	}
}

/*
 * This is the externally visible routine for
 * inode memory management.
 */
void free_inode_memory(int goal)
{
	spin_lock(&inode_lock);
	free_inodes();
	spin_unlock(&inode_lock);
}


/*
 * This is called with the spinlock held, but releases
 * the lock when freeing or allocating inodes.
 * Look out! This returns with the inode lock held if
 * it got an inode..
 *
 * We do inode allocations two pages at a time to reduce
 * fragmentation.
 */
#define INODE_PAGE_ORDER	1
#define INODE_ALLOCATION_SIZE	(PAGE_SIZE << INODE_PAGE_ORDER)
#define INODES_PER_ALLOCATION	(INODE_ALLOCATION_SIZE/sizeof(struct inode))

static struct inode * grow_inodes(void)
{
	struct inode * inode;

	/*
	 * Check whether to restock the unused list.