xattr.c

linux 内核源代码

/*
 * linux/fs/ext2/xattr.c
 *
 * Copyright (C) 2001-2003 Andreas Gruenbacher <[e-mail]>
 *
 * Fix by Harrison Xing <[e-mail]>.
 * Extended attributes for symlinks and special files added per
 *  suggestion of Luka Renko <[e-mail]>.
 * xattr consolidation Copyright (c) 2004 James Morris <[e-mail]>,
 *  Red Hat Inc.
 *
 */

/*
 * Extended attributes are stored on disk blocks allocated outside of
 * any inode. The i_file_acl field is then made to point to this allocated
 * block. If all extended attributes of an inode are identical, these
 * inodes may share the same extended attribute block. Such situations
 * are automatically detected by keeping a cache of recent attribute block
 * numbers and hashes over the block's contents in memory.
 *
 *
 * Extended attribute block layout:
 *
 *   +------------------+
 *   | header           |
 *   | entry 1          | |
 *   | entry 2          | | growing downwards
 *   | entry 3          | v
 *   | four null bytes  |
 *   | . . .            |
 *   | value 1          | ^
 *   | value 3          | | growing upwards
 *   | value 2          | |
 *   +------------------+
 *
 * The block header is followed by multiple entry descriptors. These entry
 * descriptors are variable in size, and alligned to EXT2_XATTR_PAD
 * byte boundaries. The entry descriptors are sorted by attribute name,
 * so that two extended attribute blocks can be compared efficiently.
 *
 * Attribute values are aligned to the end of the block, stored in
 * no specific order. They are also padded to EXT2_XATTR_PAD byte
 * boundaries. No additional gaps are left between them.
 *
 * Locking strategy
 * ----------------
 * EXT2_I(inode)->i_file_acl is protected by EXT2_I(inode)->xattr_sem.
 * EA blocks are only changed if they are exclusive to an inode, so
 * holding xattr_sem also means that nothing but the EA block's reference
 * count will change. Multiple writers to an EA block are synchronized
 * by the bh lock. No more than a single bh lock is held at any time
 * to avoid deadlocks.
 */

#include <linux/buffer_head.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/slab.h>
#include <linux/mbcache.h>
#include <linux/quotaops.h>
#include <linux/rwsem.h>
#include "ext2.h"
#include "xattr.h"
#include "acl.h"

#define HDR(bh) ((struct ext2_xattr_header *)((bh)->b_data))
#define ENTRY(ptr) ((struct ext2_xattr_entry *)(ptr))
#define FIRST_ENTRY(bh) ENTRY(HDR(bh)+1)
#define IS_LAST_ENTRY(entry) (*(__u32 *)(entry) == 0)

#ifdef EXT2_XATTR_DEBUG
# define ea_idebug(inode, f...) do { \
		printk(KERN_DEBUG "inode %s:%ld: ", \
			inode->i_sb->s_id, inode->i_ino); \
		printk(f); \
		printk("\n"); \
	} while (0)
# define ea_bdebug(bh, f...) do { \
		char b[BDEVNAME_SIZE]; \
		printk(KERN_DEBUG "block %s:%lu: ", \
			bdevname(bh->b_bdev, b), \
			(unsigned long) bh->b_blocknr); \
		printk(f); \
		printk("\n"); \
	} while (0)
#else
# define ea_idebug(f...)
# define ea_bdebug(f...)
#endif

static int ext2_xattr_set2(struct inode *, struct buffer_head *,
			   struct ext2_xattr_header *);

static int ext2_xattr_cache_insert(struct buffer_head *);
static struct buffer_head *ext2_xattr_cache_find(struct inode *,
						 struct ext2_xattr_header *);
static void ext2_xattr_rehash(struct ext2_xattr_header *,
			      struct ext2_xattr_entry *);

static struct mb_cache *ext2_xattr_cache;

static struct xattr_handler *ext2_xattr_handler_map[] = {
	[EXT2_XATTR_INDEX_USER]		     = &ext2_xattr_user_handler,
#ifdef CONFIG_EXT2_FS_POSIX_ACL
	[EXT2_XATTR_INDEX_POSIX_ACL_ACCESS]  = &ext2_xattr_acl_access_handler,
	[EXT2_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext2_xattr_acl_default_handler,
#endif
	[EXT2_XATTR_INDEX_TRUSTED]	     = &ext2_xattr_trusted_handler,
#ifdef CONFIG_EXT2_FS_SECURITY
	[EXT2_XATTR_INDEX_SECURITY]	     = &ext2_xattr_security_handler,
#endif
};

struct xattr_handler *ext2_xattr_handlers[] = {
	&ext2_xattr_user_handler,
	&ext2_xattr_trusted_handler,
#ifdef CONFIG_EXT2_FS_POSIX_ACL
	&ext2_xattr_acl_access_handler,
	&ext2_xattr_acl_default_handler,
#endif
#ifdef CONFIG_EXT2_FS_SECURITY
	&ext2_xattr_security_handler,
#endif
	NULL
};

static inline struct xattr_handler *
ext2_xattr_handler(int name_index)
{
	struct xattr_handler *handler = NULL;

	if (name_index > 0 && name_index < ARRAY_SIZE(ext2_xattr_handler_map))
		handler = ext2_xattr_handler_map[name_index];
	return handler;
}

/*
 * ext2_xattr_get()
 *
 * Copy an extended attribute into the buffer
 * provided, or compute the buffer size required.
 * Buffer is NULL to compute the size of the buffer required.
 *
 * Returns a negative error number on failure, or the number of bytes
 * used / required on success.
 */
int
ext2_xattr_get(struct inode *inode, int name_index, const char *name,
	       void *buffer, size_t buffer_size)
{
	struct buffer_head *bh = NULL;
	struct ext2_xattr_entry *entry;
	size_t name_len, size;
	char *end;
	int error;

	ea_idebug(inode, "name=%d.%s, buffer=%p, buffer_size=%ld",
		  name_index, name, buffer, (long)buffer_size);

	if (name == NULL)
		return -EINVAL;
	down_read(&EXT2_I(inode)->xattr_sem);
	error = -ENODATA;
	if (!EXT2_I(inode)->i_file_acl)
		goto cleanup;
	ea_idebug(inode, "reading block %d", EXT2_I(inode)->i_file_acl);
	bh = sb_bread(inode->i_sb, EXT2_I(inode)->i_file_acl);
	error = -EIO;
	if (!bh)
		goto cleanup;
	ea_bdebug(bh, "b_count=%d, refcount=%d",
		atomic_read(&(bh->b_count)), le32_to_cpu(HDR(bh)->h_refcount));
	end = bh->b_data + bh->b_size;
	if (HDR(bh)->h_magic != cpu_to_le32(EXT2_XATTR_MAGIC) ||
	    HDR(bh)->h_blocks != cpu_to_le32(1)) {
bad_block:	ext2_error(inode->i_sb, "ext2_xattr_get",
			"inode %ld: bad block %d", inode->i_ino,
			EXT2_I(inode)->i_file_acl);
		error = -EIO;
		goto cleanup;
	}
	/* find named attribute */
	name_len = strlen(name);

	error = -ERANGE;
	if (name_len > 255)
		goto cleanup;
	entry = FIRST_ENTRY(bh);
	while (!IS_LAST_ENTRY(entry)) {
		struct ext2_xattr_entry *next =
			EXT2_XATTR_NEXT(entry);
		if ((char *)next >= end)
			goto bad_block;
		if (name_index == entry->e_name_index &&
		    name_len == entry->e_name_len &&
		    memcmp(name, entry->e_name, name_len) == 0)
			goto found;
		entry = next;
	}
	/* Check the remaining name entries */
	while (!IS_LAST_ENTRY(entry)) {
		struct ext2_xattr_entry *next =
			EXT2_XATTR_NEXT(entry);
		if ((char *)next >= end)
			goto bad_block;
		entry = next;
	}
	if (ext2_xattr_cache_insert(bh))
		ea_idebug(inode, "cache insert failed");
	error = -ENODATA;
	goto cleanup;
found:
	/* check the buffer size */
	if (entry->e_value_block != 0)
		goto bad_block;
	size = le32_to_cpu(entry->e_value_size);
	if (size > inode->i_sb->s_blocksize ||
	    le16_to_cpu(entry->e_value_offs) + size > inode->i_sb->s_blocksize)
		goto bad_block;

	if (ext2_xattr_cache_insert(bh))
		ea_idebug(inode, "cache insert failed");
	if (buffer) {
		error = -ERANGE;
		if (size > buffer_size)
			goto cleanup;
		/* return value of attribute */
		memcpy(buffer, bh->b_data + le16_to_cpu(entry->e_value_offs),
			size);
	}
	error = size;

cleanup:
	brelse(bh);
	up_read(&EXT2_I(inode)->xattr_sem);

	return error;
}

/*
 * ext2_xattr_list()
 *
 * Copy a list of attribute names into the buffer
 * provided, or compute the buffer size required.
 * Buffer is NULL to compute the size of the buffer required.
 *
 * Returns a negative error number on failure, or the number of bytes
 * used / required on success.
 */
static int
ext2_xattr_list(struct inode *inode, char *buffer, size_t buffer_size)
{
	struct buffer_head *bh = NULL;
	struct ext2_xattr_entry *entry;
	char *end;
	size_t rest = buffer_size;
	int error;

	ea_idebug(inode, "buffer=%p, buffer_size=%ld",
		  buffer, (long)buffer_size);

	down_read(&EXT2_I(inode)->xattr_sem);
	error = 0;
	if (!EXT2_I(inode)->i_file_acl)
		goto cleanup;
	ea_idebug(inode, "reading block %d", EXT2_I(inode)->i_file_acl);
	bh = sb_bread(inode->i_sb, EXT2_I(inode)->i_file_acl);
	error = -EIO;
	if (!bh)
		goto cleanup;
	ea_bdebug(bh, "b_count=%d, refcount=%d",
		atomic_read(&(bh->b_count)), le32_to_cpu(HDR(bh)->h_refcount));
	end = bh->b_data + bh->b_size;
	if (HDR(bh)->h_magic != cpu_to_le32(EXT2_XATTR_MAGIC) ||
	    HDR(bh)->h_blocks != cpu_to_le32(1)) {
bad_block:	ext2_error(inode->i_sb, "ext2_xattr_list",
			"inode %ld: bad block %d", inode->i_ino,
			EXT2_I(inode)->i_file_acl);
		error = -EIO;
		goto cleanup;
	}

	/* check the on-disk data structure */
	entry = FIRST_ENTRY(bh);
	while (!IS_LAST_ENTRY(entry)) {
		struct ext2_xattr_entry *next = EXT2_XATTR_NEXT(entry);

		if ((char *)next >= end)
			goto bad_block;
		entry = next;
	}
	if (ext2_xattr_cache_insert(bh))
		ea_idebug(inode, "cache insert failed");

	/* list the attribute names */
	for (entry = FIRST_ENTRY(bh); !IS_LAST_ENTRY(entry);
	     entry = EXT2_XATTR_NEXT(entry)) {
		struct xattr_handler *handler =
			ext2_xattr_handler(entry->e_name_index);

		if (handler) {
			size_t size = handler->list(inode, buffer, rest,
						    entry->e_name,
						    entry->e_name_len);
			if (buffer) {
				if (size > rest) {
					error = -ERANGE;
					goto cleanup;
				}
				buffer += size;
			}
			rest -= size;
		}
	}
	error = buffer_size - rest;  /* total size */

cleanup:
	brelse(bh);
	up_read(&EXT2_I(inode)->xattr_sem);

	return error;
}

/*
 * Inode operation listxattr()
 *
 * dentry->d_inode->i_mutex: don't care
 */
ssize_t
ext2_listxattr(struct dentry *dentry, char *buffer, size_t size)
{
	return ext2_xattr_list(dentry->d_inode, buffer, size);
}

/*
 * If the EXT2_FEATURE_COMPAT_EXT_ATTR feature of this file system is
 * not set, set it.
 */
static void ext2_xattr_update_super_block(struct super_block *sb)
{
	if (EXT2_HAS_COMPAT_FEATURE(sb, EXT2_FEATURE_COMPAT_EXT_ATTR))
		return;

	EXT2_SET_COMPAT_FEATURE(sb, EXT2_FEATURE_COMPAT_EXT_ATTR);
	sb->s_dirt = 1;
	mark_buffer_dirty(EXT2_SB(sb)->s_sbh);
}

/*
 * ext2_xattr_set()
 *
 * Create, replace or remove an extended attribute for this inode. Buffer
 * is NULL to remove an existing extended attribute, and non-NULL to
 * either replace an existing extended attribute, or create a new extended
 * attribute. The flags XATTR_REPLACE and XATTR_CREATE
 * specify that an extended attribute must exist and must not exist
 * previous to the call, respectively.
 *
 * Returns 0, or a negative error number on failure.
 */
int
ext2_xattr_set(struct inode *inode, int name_index, const char *name,
	       const void *value, size_t value_len, int flags)
{
	struct super_block *sb = inode->i_sb;
	struct buffer_head *bh = NULL;
	struct ext2_xattr_header *header = NULL;
	struct ext2_xattr_entry *here, *last;
	size_t name_len, free, min_offs = sb->s_blocksize;
	int not_found = 1, error;
	char *end;
	
	/*
	 * header -- Points either into bh, or to a temporarily
	 *           allocated buffer.
	 * here -- The named entry found, or the place for inserting, within
	 *         the block pointed to by header.
	 * last -- Points right after the last named entry within the block
	 *         pointed to by header.
	 * min_offs -- The offset of the first value (values are aligned
	 *             towards the end of the block).
	 * end -- Points right after the block pointed to by header.
	 */
	
	ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld",
		  name_index, name, value, (long)value_len);

	if (value == NULL)
		value_len = 0;
	if (name == NULL)
		return -EINVAL;
	name_len = strlen(name);
	if (name_len > 255 || value_len > sb->s_blocksize)
		return -ERANGE;
	down_write(&EXT2_I(inode)->xattr_sem);
	if (EXT2_I(inode)->i_file_acl) {
		/* The inode already has an extended attribute block. */
		bh = sb_bread(sb, EXT2_I(inode)->i_file_acl);
		error = -EIO;
		if (!bh)
			goto cleanup;
		ea_bdebug(bh, "b_count=%d, refcount=%d",
			atomic_read(&(bh->b_count)),
			le32_to_cpu(HDR(bh)->h_refcount));
		header = HDR(bh);
		end = bh->b_data + bh->b_size;
		if (header->h_magic != cpu_to_le32(EXT2_XATTR_MAGIC) ||
		    header->h_blocks != cpu_to_le32(1)) {
bad_block:		ext2_error(sb, "ext2_xattr_set",
				"inode %ld: bad block %d", inode->i_ino, 
				   EXT2_I(inode)->i_file_acl);
			error = -EIO;
			goto cleanup;
		}
		/* Find the named attribute. */
		here = FIRST_ENTRY(bh);
		while (!IS_LAST_ENTRY(here)) {
			struct ext2_xattr_entry *next = EXT2_XATTR_NEXT(here);
			if ((char *)next >= end)
				goto bad_block;
			if (!here->e_value_block && here->e_value_size) {
				size_t offs = le16_to_cpu(here->e_value_offs);
				if (offs < min_offs)
					min_offs = offs;
			}
			not_found = name_index - here->e_name_index;
			if (!not_found)
				not_found = name_len - here->e_name_len;
			if (!not_found)
				not_found = memcmp(name, here->e_name,name_len);
			if (not_found <= 0)
				break;
			here = next;
		}
		last = here;
		/* We still need to compute min_offs and last. */
		while (!IS_LAST_ENTRY(last)) {
			struct ext2_xattr_entry *next = EXT2_XATTR_NEXT(last);
			if ((char *)next >= end)
				goto bad_block;
			if (!last->e_value_block && last->e_value_size) {
				size_t offs = le16_to_cpu(last->e_value_offs);
				if (offs < min_offs)
					min_offs = offs;
			}
			last = next;
		}

		/* Check whether we have enough space left. */
		free = min_offs - ((char*)last - (char*)header) - sizeof(__u32);
	} else {
		/* We will use a new extended attribute block. */
		free = sb->s_blocksize -
			sizeof(struct ext2_xattr_header) - sizeof(__u32);
		here = last = NULL;  /* avoid gcc uninitialized warning. */
	}

	if (not_found) {
		/* Request to remove a nonexistent attribute? */
		error = -ENODATA;
		if (flags & XATTR_REPLACE)
			goto cleanup;
		error = 0;
		if (value == NULL)
			goto cleanup;
	} else {
		/* Request to create an existing attribute? */
		error = -EEXIST;
		if (flags & XATTR_CREATE)
			goto cleanup;
		if (!here->e_value_block && here->e_value_size) {
			size_t size = le32_to_cpu(here->e_value_size);

			if (le16_to_cpu(here->e_value_offs) + size > 
			    sb->s_blocksize || size > sb->s_blocksize)
				goto bad_block;
			free += EXT2_XATTR_SIZE(size);
		}
		free += EXT2_XATTR_LEN(name_len);
	}
	error = -ENOSPC;
	if (free < EXT2_XATTR_LEN(name_len) + EXT2_XATTR_SIZE(value_len))
		goto cleanup;

	/* Here we know that we can set the new attribute. */

	if (header) {
		struct mb_cache_entry *ce;

		/* assert(header == HDR(bh)); */
		ce = mb_cache_entry_get(ext2_xattr_cache, bh->b_bdev,
					bh->b_blocknr);
		lock_buffer(bh);
		if (header->h_refcount == cpu_to_le32(1)) {
			ea_bdebug(bh, "modifying in-place");
			if (ce)
				mb_cache_entry_free(ce);
			/* keep the buffer locked while modifying it. */
		} else {
			int offset;

			if (ce)
				mb_cache_entry_release(ce);
			unlock_buffer(bh);
			ea_bdebug(bh, "cloning");
			header = kmalloc(bh->b_size, GFP_KERNEL);
			error = -ENOMEM;
			if (header == NULL)
				goto cleanup;
			memcpy(header, HDR(bh), bh->b_size);
			header->h_refcount = cpu_to_le32(1);

			offset = (char *)here - bh->b_data;
			here = ENTRY((char *)header + offset);
			offset = (char *)last - bh->b_data;
			last = ENTRY((char *)header + offset);
		}
	} else {