#include <linux/time.h>
#include <linux/fs.h>
#include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
#include <linux/stat.h>
#include <linux/string.h>
#include <linux/quotaops.h>
#include <linux/bitops.h>
#include <linux/blkdev.h>
#include <asm/byteorder.h>
-
+#include "ext4.h"
+#include "ext4_jbd2.h"
#include "xattr.h"
#include "acl.h"
#include "group.h"
/* If checksum is bad mark all blocks and inodes use to prevent
* allocation, essentially implementing a per-group read-only flag. */
if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) {
- ext4_error(sb, __FUNCTION__, "Checksum bad for group %lu\n",
+ ext4_error(sb, __func__, "Checksum bad for group %lu\n",
block_group);
gdp->bg_free_blocks_count = 0;
gdp->bg_free_inodes_count = 0;
* Return buffer_head of bitmap on success or NULL.
*/
static struct buffer_head *
-read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
+ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
{
struct ext4_group_desc *desc;
struct buffer_head *bh = NULL;
+ ext4_fsblk_t bitmap_blk;
desc = ext4_get_group_desc(sb, block_group, NULL);
if (!desc)
- goto error_out;
+ return NULL;
+ bitmap_blk = ext4_inode_bitmap(sb, desc);
+ bh = sb_getblk(sb, bitmap_blk);
+ if (unlikely(!bh)) {
+ ext4_error(sb, __func__,
+ "Cannot read inode bitmap - "
+ "block_group = %lu, inode_bitmap = %llu",
+ block_group, bitmap_blk);
+ return NULL;
+ }
+ if (buffer_uptodate(bh) &&
+ !(desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)))
+ return bh;
+
+ lock_buffer(bh);
+ spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
- bh = sb_getblk(sb, ext4_inode_bitmap(sb, desc));
- if (!buffer_uptodate(bh)) {
- lock_buffer(bh);
- if (!buffer_uptodate(bh)) {
- ext4_init_inode_bitmap(sb, bh, block_group,
- desc);
- set_buffer_uptodate(bh);
- }
- unlock_buffer(bh);
- }
- } else {
- bh = sb_bread(sb, ext4_inode_bitmap(sb, desc));
+ ext4_init_inode_bitmap(sb, bh, block_group, desc);
+ set_buffer_uptodate(bh);
+ unlock_buffer(bh);
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
+ return bh;
}
- if (!bh)
- ext4_error(sb, "read_inode_bitmap",
+ spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
+ if (bh_submit_read(bh) < 0) {
+ put_bh(bh);
+ ext4_error(sb, __func__,
"Cannot read inode bitmap - "
"block_group = %lu, inode_bitmap = %llu",
- block_group, ext4_inode_bitmap(sb, desc));
-error_out:
+ block_group, bitmap_blk);
+ return NULL;
+ }
return bh;
}
* though), and then we'd have two inodes sharing the
* same inode number and space on the harddisk.
*/
-void ext4_free_inode (handle_t *handle, struct inode * inode)
+void ext4_free_inode(handle_t *handle, struct inode *inode)
{
- struct super_block * sb = inode->i_sb;
+ struct super_block *sb = inode->i_sb;
int is_directory;
unsigned long ino;
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *bh2;
ext4_group_t block_group;
unsigned long bit;
- struct ext4_group_desc * gdp;
- struct ext4_super_block * es;
+ struct ext4_group_desc *gdp;
+ struct ext4_super_block *es;
struct ext4_sb_info *sbi;
int fatal = 0, err;
+ ext4_group_t flex_group;
if (atomic_read(&inode->i_count) > 1) {
- printk ("ext4_free_inode: inode has count=%d\n",
- atomic_read(&inode->i_count));
+ printk(KERN_ERR "ext4_free_inode: inode has count=%d\n",
+ atomic_read(&inode->i_count));
return;
}
if (inode->i_nlink) {
- printk ("ext4_free_inode: inode has nlink=%d\n",
- inode->i_nlink);
+ printk(KERN_ERR "ext4_free_inode: inode has nlink=%d\n",
+ inode->i_nlink);
return;
}
if (!sb) {
- printk("ext4_free_inode: inode on nonexistent device\n");
+ printk(KERN_ERR "ext4_free_inode: inode on "
+ "nonexistent device\n");
return;
}
sbi = EXT4_SB(sb);
ino = inode->i_ino;
- ext4_debug ("freeing inode %lu\n", ino);
+ ext4_debug("freeing inode %lu\n", ino);
/*
* Note: we must free any quota before locking the superblock,
is_directory = S_ISDIR(inode->i_mode);
/* Do this BEFORE marking the inode not in use or returning an error */
- clear_inode (inode);
+ clear_inode(inode);
es = EXT4_SB(sb)->s_es;
if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
- ext4_error (sb, "ext4_free_inode",
- "reserved or nonexistent inode %lu", ino);
+ ext4_error(sb, "ext4_free_inode",
+ "reserved or nonexistent inode %lu", ino);
goto error_return;
}
block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
- bitmap_bh = read_inode_bitmap(sb, block_group);
+ bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
if (!bitmap_bh)
goto error_return;
/* Ok, now we can actually update the inode bitmaps.. */
if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
bit, bitmap_bh->b_data))
- ext4_error (sb, "ext4_free_inode",
- "bit already cleared for inode %lu", ino);
+ ext4_error(sb, "ext4_free_inode",
+ "bit already cleared for inode %lu", ino);
else {
- gdp = ext4_get_group_desc (sb, block_group, &bh2);
+ gdp = ext4_get_group_desc(sb, block_group, &bh2);
BUFFER_TRACE(bh2, "get_write_access");
fatal = ext4_journal_get_write_access(handle, bh2);
if (gdp) {
spin_lock(sb_bgl_lock(sbi, block_group));
- gdp->bg_free_inodes_count = cpu_to_le16(
- le16_to_cpu(gdp->bg_free_inodes_count) + 1);
+ le16_add_cpu(&gdp->bg_free_inodes_count, 1);
if (is_directory)
- gdp->bg_used_dirs_count = cpu_to_le16(
- le16_to_cpu(gdp->bg_used_dirs_count) - 1);
+ le16_add_cpu(&gdp->bg_used_dirs_count, -1);
gdp->bg_checksum = ext4_group_desc_csum(sbi,
block_group, gdp);
spin_unlock(sb_bgl_lock(sbi, block_group));
if (is_directory)
percpu_counter_dec(&sbi->s_dirs_counter);
+ if (sbi->s_log_groups_per_flex) {
+ flex_group = ext4_flex_group(sbi, block_group);
+ spin_lock(sb_bgl_lock(sbi, flex_group));
+ sbi->s_flex_groups[flex_group].free_inodes++;
+ spin_unlock(sb_bgl_lock(sbi, flex_group));
+ }
}
BUFFER_TRACE(bh2, "call ext4_journal_dirty_metadata");
err = ext4_journal_dirty_metadata(handle, bh2);
avefreei = freei / ngroups;
for (group = 0; group < ngroups; group++) {
- desc = ext4_get_group_desc (sb, group, NULL);
+ desc = ext4_get_group_desc(sb, group, NULL);
if (!desc || !desc->bg_free_inodes_count)
continue;
if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei)
return ret;
}
+#define free_block_ratio 10
+
+static int find_group_flex(struct super_block *sb, struct inode *parent,
+ ext4_group_t *best_group)
+{
+ struct ext4_sb_info *sbi = EXT4_SB(sb);
+ struct ext4_group_desc *desc;
+ struct buffer_head *bh;
+ struct flex_groups *flex_group = sbi->s_flex_groups;
+ ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
+ ext4_group_t parent_fbg_group = ext4_flex_group(sbi, parent_group);
+ ext4_group_t ngroups = sbi->s_groups_count;
+ int flex_size = ext4_flex_bg_size(sbi);
+ ext4_group_t best_flex = parent_fbg_group;
+ int blocks_per_flex = sbi->s_blocks_per_group * flex_size;
+ int flexbg_free_blocks;
+ int flex_freeb_ratio;
+ ext4_group_t n_fbg_groups;
+ ext4_group_t i;
+
+ n_fbg_groups = (sbi->s_groups_count + flex_size - 1) >>
+ sbi->s_log_groups_per_flex;
+
+find_close_to_parent:
+ flexbg_free_blocks = flex_group[best_flex].free_blocks;
+ flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex;
+ if (flex_group[best_flex].free_inodes &&
+ flex_freeb_ratio > free_block_ratio)
+ goto found_flexbg;
+
+ if (best_flex && best_flex == parent_fbg_group) {
+ best_flex--;
+ goto find_close_to_parent;
+ }
+
+ for (i = 0; i < n_fbg_groups; i++) {
+ if (i == parent_fbg_group || i == parent_fbg_group - 1)
+ continue;
+
+ flexbg_free_blocks = flex_group[i].free_blocks;
+ flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex;
+
+ if (flex_freeb_ratio > free_block_ratio &&
+ flex_group[i].free_inodes) {
+ best_flex = i;
+ goto found_flexbg;
+ }
+
+ if (flex_group[best_flex].free_inodes == 0 ||
+ (flex_group[i].free_blocks >
+ flex_group[best_flex].free_blocks &&
+ flex_group[i].free_inodes))
+ best_flex = i;
+ }
+
+ if (!flex_group[best_flex].free_inodes ||
+ !flex_group[best_flex].free_blocks)
+ return -1;
+
+found_flexbg:
+ for (i = best_flex * flex_size; i < ngroups &&
+ i < (best_flex + 1) * flex_size; i++) {
+ desc = ext4_get_group_desc(sb, i, &bh);
+ if (le16_to_cpu(desc->bg_free_inodes_count)) {
+ *best_group = i;
+ goto out;
+ }
+ }
+
+ return -1;
+out:
+ return 0;
+}
+
/*
* Orlov's allocator for directories.
*
* it has too few free inodes left (min_inodes) or
* it has too few free blocks left (min_blocks) or
* it's already running too large debt (max_debt).
- * Parent's group is prefered, if it doesn't satisfy these
+ * Parent's group is preferred, if it doesn't satisfy these
* conditions we search cyclically through the rest. If none
* of the groups look good we just look for a group with more
* free inodes than average (starting at parent's group).
* For other inodes, search forward from the parent directory's block
* group to find a free inode.
*/
-struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
+struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode)
{
struct super_block *sb;
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *bh2;
ext4_group_t group = 0;
unsigned long ino = 0;
- struct inode * inode;
- struct ext4_group_desc * gdp = NULL;
- struct ext4_super_block * es;
+ struct inode *inode;
+ struct ext4_group_desc *gdp = NULL;
+ struct ext4_super_block *es;
struct ext4_inode_info *ei;
struct ext4_sb_info *sbi;
int ret2, err = 0;
struct inode *ret;
ext4_group_t i;
int free = 0;
+ ext4_group_t flex_group;
/* Cannot create files in a deleted directory */
if (!dir || !dir->i_nlink)
sbi = EXT4_SB(sb);
es = sbi->s_es;
+
+ if (sbi->s_log_groups_per_flex) {
+ ret2 = find_group_flex(sb, dir, &group);
+ goto got_group;
+ }
+
if (S_ISDIR(mode)) {
- if (test_opt (sb, OLDALLOC))
+ if (test_opt(sb, OLDALLOC))
ret2 = find_group_dir(sb, dir, &group);
else
ret2 = find_group_orlov(sb, dir, &group);
} else
ret2 = find_group_other(sb, dir, &group);
+got_group:
err = -ENOSPC;
if (ret2 == -1)
goto out;
goto fail;
brelse(bitmap_bh);
- bitmap_bh = read_inode_bitmap(sb, group);
+ bitmap_bh = ext4_read_inode_bitmap(sb, group);
if (!bitmap_bh)
goto fail;
ino++;
if ((group == 0 && ino < EXT4_FIRST_INO(sb)) ||
ino > EXT4_INODES_PER_GROUP(sb)) {
- ext4_error(sb, __FUNCTION__,
+ ext4_error(sb, __func__,
"reserved inode or inode > inodes count - "
"block_group = %lu, inode=%lu", group,
ino + group * EXT4_INODES_PER_GROUP(sb));
/* We may have to initialize the block bitmap if it isn't already */
if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM) &&
gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
- struct buffer_head *block_bh = read_block_bitmap(sb, group);
+ struct buffer_head *block_bh = ext4_read_block_bitmap(sb, group);
BUFFER_TRACE(block_bh, "get block bitmap access");
err = ext4_journal_get_write_access(handle, block_bh);
gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
free = ext4_free_blocks_after_init(sb, group, gdp);
gdp->bg_free_blocks_count = cpu_to_le16(free);
+ gdp->bg_checksum = ext4_group_desc_csum(sbi, group,
+ gdp);
}
spin_unlock(sb_bgl_lock(sbi, group));
/* When marking the block group with
* ~EXT4_BG_INODE_UNINIT we don't want to depend
- * on the value of bg_itable_unsed even though
+ * on the value of bg_itable_unused even though
* mke2fs could have initialized the same for us.
* Instead we calculated the value below
*/
cpu_to_le16(EXT4_INODES_PER_GROUP(sb) - ino);
}
- gdp->bg_free_inodes_count =
- cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1);
+ le16_add_cpu(&gdp->bg_free_inodes_count, -1);
if (S_ISDIR(mode)) {
- gdp->bg_used_dirs_count =
- cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1);
+ le16_add_cpu(&gdp->bg_used_dirs_count, 1);
}
gdp->bg_checksum = ext4_group_desc_csum(sbi, group, gdp);
spin_unlock(sb_bgl_lock(sbi, group));
percpu_counter_inc(&sbi->s_dirs_counter);
sb->s_dirt = 1;
- inode->i_uid = current->fsuid;
- if (test_opt (sb, GRPID))
+ if (sbi->s_log_groups_per_flex) {
+ flex_group = ext4_flex_group(sbi, group);
+ spin_lock(sb_bgl_lock(sbi, flex_group));
+ sbi->s_flex_groups[flex_group].free_inodes--;
+ spin_unlock(sb_bgl_lock(sbi, flex_group));
+ }
+
+ inode->i_uid = current_fsuid();
+ if (test_opt(sb, GRPID))
inode->i_gid = dir->i_gid;
else if (dir->i_mode & S_ISGID) {
inode->i_gid = dir->i_gid;
if (S_ISDIR(mode))
mode |= S_ISGID;
} else
- inode->i_gid = current->fsgid;
+ inode->i_gid = current_fsgid();
inode->i_mode = mode;
inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb);
ei->i_dir_start_lookup = 0;
ei->i_disksize = 0;
- ei->i_flags = EXT4_I(dir)->i_flags & ~EXT4_INDEX_FL;
+ /*
+ * Don't inherit extent flag from directory. We set extent flag on
+ * newly created directory and file only if -o extent mount option is
+ * specified
+ */
+ ei->i_flags = EXT4_I(dir)->i_flags & ~(EXT4_INDEX_FL|EXT4_EXTENTS_FL);
if (S_ISLNK(mode))
ei->i_flags &= ~(EXT4_IMMUTABLE_FL|EXT4_APPEND_FL);
/* dirsync only applies to directories */
ei->i_flags &= ~EXT4_DIRSYNC_FL;
ei->i_file_acl = 0;
ei->i_dtime = 0;
- ei->i_block_alloc_info = NULL;
ei->i_block_group = group;
ext4_set_inode_flags(inode);
if (IS_DIRSYNC(inode))
handle->h_sync = 1;
- insert_inode_hash(inode);
+ if (insert_inode_locked(inode) < 0) {
+ err = -EINVAL;
+ goto fail_drop;
+ }
spin_lock(&sbi->s_next_gen_lock);
inode->i_generation = sbi->s_next_generation++;
spin_unlock(&sbi->s_next_gen_lock);
ei->i_extra_isize = EXT4_SB(sb)->s_want_extra_isize;
ret = inode;
- if(DQUOT_ALLOC_INODE(inode)) {
+ if (DQUOT_ALLOC_INODE(inode)) {
err = -EDQUOT;
goto fail_drop;
}
if (err)
goto fail_free_drop;
- err = ext4_init_security(handle,inode, dir);
+ err = ext4_init_security(handle, inode, dir);
if (err)
goto fail_free_drop;
+ if (test_opt(sb, EXTENTS)) {
+ /* set extent flag only for directory, file and normal symlink*/
+ if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) {
+ EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL;
+ ext4_ext_tree_init(handle, inode);
+ }
+ }
+
err = ext4_mark_inode_dirty(handle, inode);
if (err) {
ext4_std_error(sb, err);
goto fail_free_drop;
}
- if (test_opt(sb, EXTENTS)) {
- EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL;
- ext4_ext_tree_init(handle, inode);
- err = ext4_update_incompat_feature(handle, sb,
- EXT4_FEATURE_INCOMPAT_EXTENTS);
- if (err)
- goto fail;
- }
ext4_debug("allocating inode %lu\n", inode->i_ino);
goto really_out;
DQUOT_DROP(inode);
inode->i_flags |= S_NOQUOTA;
inode->i_nlink = 0;
+ unlock_new_inode(inode);
iput(inode);
brelse(bitmap_bh);
return ERR_PTR(err);
/* Error cases - e2fsck has already cleaned up for us */
if (ino > max_ino) {
- ext4_warning(sb, __FUNCTION__,
+ ext4_warning(sb, __func__,
"bad orphan ino %lu! e2fsck was run?", ino);
goto error;
}
block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
- bitmap_bh = read_inode_bitmap(sb, block_group);
+ bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
if (!bitmap_bh) {
- ext4_warning(sb, __FUNCTION__,
+ ext4_warning(sb, __func__,
"inode bitmap error for orphan %lu", ino);
goto error;
}
if (IS_ERR(inode))
goto iget_failed;
+ /*
+ * If the orphans has i_nlinks > 0 then it should be able to be
+ * truncated, otherwise it won't be removed from the orphan list
+ * during processing and an infinite loop will result.
+ */
+ if (inode->i_nlink && !ext4_can_truncate(inode))
+ goto bad_orphan;
+
if (NEXT_ORPHAN(inode) > max_ino)
goto bad_orphan;
brelse(bitmap_bh);
err = PTR_ERR(inode);
inode = NULL;
bad_orphan:
- ext4_warning(sb, __FUNCTION__,
+ ext4_warning(sb, __func__,
"bad orphan inode %lu! e2fsck was run?", ino);
printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n",
bit, (unsigned long long)bitmap_bh->b_blocknr,
printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n",
NEXT_ORPHAN(inode));
printk(KERN_NOTICE "max_ino=%lu\n", max_ino);
+ printk(KERN_NOTICE "i_nlink=%u\n", inode->i_nlink);
/* Avoid freeing blocks if we got a bad deleted inode */
if (inode->i_nlink == 0)
inode->i_blocks = 0;
return ERR_PTR(err);
}
-unsigned long ext4_count_free_inodes (struct super_block * sb)
+unsigned long ext4_count_free_inodes(struct super_block *sb)
{
unsigned long desc_count;
struct ext4_group_desc *gdp;
bitmap_count = 0;
gdp = NULL;
for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
- gdp = ext4_get_group_desc (sb, i, NULL);
+ gdp = ext4_get_group_desc(sb, i, NULL);
if (!gdp)
continue;
desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
brelse(bitmap_bh);
- bitmap_bh = read_inode_bitmap(sb, i);
+ bitmap_bh = ext4_read_inode_bitmap(sb, i);
if (!bitmap_bh)
continue;
bitmap_count += x;
}
brelse(bitmap_bh);
- printk("ext4_count_free_inodes: stored = %u, computed = %lu, %lu\n",
- le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
+ printk(KERN_DEBUG "ext4_count_free_inodes: "
+ "stored = %u, computed = %lu, %lu\n",
+ le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
return desc_count;
#else
desc_count = 0;
for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
- gdp = ext4_get_group_desc (sb, i, NULL);
+ gdp = ext4_get_group_desc(sb, i, NULL);
if (!gdp)
continue;
desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
}
/* Called at mount-time, super-block is locked */
-unsigned long ext4_count_dirs (struct super_block * sb)
+unsigned long ext4_count_dirs(struct super_block * sb)
{
unsigned long count = 0;
ext4_group_t i;
for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
- struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL);
+ struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL);
if (!gdp)
continue;
count += le16_to_cpu(gdp->bg_used_dirs_count);