summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorStephen Rothwell <sfr@canb.auug.org.au>2008-02-28 15:40:21 +1100
committerStephen Rothwell <sfr@canb.auug.org.au>2008-02-28 15:40:21 +1100
commit486b1f6542264ab2b929cfb444971edc8da2b3e3 (patch)
tree16d28580badc5b74782ef75dc1f2f6c0cf475066
parentdfd2fa7fad31a150511833e4ee8bf9bfc295be8f (diff)
parent91f6f52c177c11c7163433da954768b6c781d9df (diff)
Merge commit 'ext4/next'
-rw-r--r--fs/ext4/balloc.c29
-rw-r--r--fs/ext4/dir.c2
-rw-r--r--fs/ext4/extents.c20
-rw-r--r--fs/ext4/file.c2
-rw-r--r--fs/ext4/fsync.c3
-rw-r--r--fs/ext4/ialloc.c108
-rw-r--r--fs/ext4/inode.c20
-rw-r--r--fs/ext4/ioctl.c12
-rw-r--r--fs/ext4/mballoc.c63
-rw-r--r--fs/ext4/namei.c9
-rw-r--r--fs/ext4/resize.c6
-rw-r--r--fs/ext4/super.c61
-rw-r--r--fs/ext4/xattr.c12
-rw-r--r--fs/jbd2/journal.c13
-rw-r--r--fs/jbd2/revoke.c4
-rw-r--r--include/linux/ext4_fs.h28
-rw-r--r--include/linux/ext4_fs_sb.h3
17 files changed, 278 insertions, 117 deletions
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 0737e05ba3dd..2f2820f88e41 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -48,7 +48,6 @@ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr,
unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
ext4_group_t block_group, struct ext4_group_desc *gdp)
{
- unsigned long start;
int bit, bit_max;
unsigned free_blocks, group_blocks;
struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -106,11 +105,12 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
free_blocks = group_blocks - bit_max;
if (bh) {
+ ext4_fsblk_t start;
+
for (bit = 0; bit < bit_max; bit++)
ext4_set_bit(bit, bh->b_data);
- start = block_group * EXT4_BLOCKS_PER_GROUP(sb) +
- le32_to_cpu(sbi->s_es->s_first_data_block);
+ start = ext4_group_first_block_no(sb, block_group);
/* Set bits for block and inode bitmaps, and inode table */
ext4_set_bit(ext4_block_bitmap(sb, gdp) - start, bh->b_data);
@@ -127,6 +127,8 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
mark_bitmap_end(group_blocks, sb->s_blocksize * 8, bh->b_data);
}
+ if (sbi->s_log_groups_per_flex)
+ return free_blocks;
return free_blocks - sbi->s_itb_per_group - 2;
}
@@ -752,13 +754,18 @@ do_more:
jbd_unlock_bh_state(bitmap_bh);
spin_lock(sb_bgl_lock(sbi, block_group));
- desc->bg_free_blocks_count =
- cpu_to_le16(le16_to_cpu(desc->bg_free_blocks_count) +
- group_freed);
+ le16_add_cpu(&desc->bg_free_blocks_count, group_freed);
desc->bg_checksum = ext4_group_desc_csum(sbi, block_group, desc);
spin_unlock(sb_bgl_lock(sbi, block_group));
percpu_counter_add(&sbi->s_freeblocks_counter, count);
+ if (sbi->s_log_groups_per_flex) {
+ ext4_group_t flex_group = ext4_flex_group(sbi, block_group);
+ spin_lock(sb_bgl_lock(sbi, flex_group));
+ sbi->s_flex_groups[flex_group].free_blocks += count;
+ spin_unlock(sb_bgl_lock(sbi, flex_group));
+ }
+
/* We dirtied the bitmap block */
BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
err = ext4_journal_dirty_metadata(handle, bitmap_bh);
@@ -1823,12 +1830,18 @@ allocated:
spin_lock(sb_bgl_lock(sbi, group_no));
if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))
gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
- gdp->bg_free_blocks_count =
- cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)-num);
+ le16_add_cpu(&gdp->bg_free_blocks_count, -num);
gdp->bg_checksum = ext4_group_desc_csum(sbi, group_no, gdp);
spin_unlock(sb_bgl_lock(sbi, group_no));
percpu_counter_sub(&sbi->s_freeblocks_counter, num);
+ if (sbi->s_log_groups_per_flex) {
+ ext4_group_t flex_group = ext4_flex_group(sbi, group_no);
+ spin_lock(sb_bgl_lock(sbi, flex_group));
+ sbi->s_flex_groups[flex_group].free_blocks -= num;
+ spin_unlock(sb_bgl_lock(sbi, flex_group));
+ }
+
BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor");
err = ext4_journal_dirty_metadata(handle, gdp_bh);
if (!fatal)
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index 2c23bade9aa6..88c97f7312be 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -42,7 +42,7 @@ const struct file_operations ext4_dir_operations = {
.llseek = generic_file_llseek,
.read = generic_read_dir,
.readdir = ext4_readdir, /* we take BKL. needed?*/
- .ioctl = ext4_ioctl, /* BKL held */
+ .unlocked_ioctl = ext4_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = ext4_compat_ioctl,
#endif
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 9ae6e67090cd..39d53152c4ac 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -614,7 +614,7 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode,
ix->ei_block = cpu_to_le32(logical);
ext4_idx_store_pblock(ix, ptr);
- curp->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(curp->p_hdr->eh_entries)+1);
+ le16_add_cpu(&curp->p_hdr->eh_entries, 1);
BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries)
> le16_to_cpu(curp->p_hdr->eh_max));
@@ -736,7 +736,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
}
if (m) {
memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m);
- neh->eh_entries = cpu_to_le16(le16_to_cpu(neh->eh_entries)+m);
+ le16_add_cpu(&neh->eh_entries, m);
}
set_buffer_uptodate(bh);
@@ -753,8 +753,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
err = ext4_ext_get_access(handle, inode, path + depth);
if (err)
goto cleanup;
- path[depth].p_hdr->eh_entries =
- cpu_to_le16(le16_to_cpu(path[depth].p_hdr->eh_entries)-m);
+ le16_add_cpu(&path[depth].p_hdr->eh_entries, -m);
err = ext4_ext_dirty(handle, inode, path + depth);
if (err)
goto cleanup;
@@ -817,8 +816,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
if (m) {
memmove(++fidx, path[i].p_idx - m,
sizeof(struct ext4_extent_idx) * m);
- neh->eh_entries =
- cpu_to_le16(le16_to_cpu(neh->eh_entries) + m);
+ le16_add_cpu(&neh->eh_entries, m);
}
set_buffer_uptodate(bh);
unlock_buffer(bh);
@@ -834,7 +832,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
err = ext4_ext_get_access(handle, inode, path + i);
if (err)
goto cleanup;
- path[i].p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path[i].p_hdr->eh_entries)-m);
+ le16_add_cpu(&path[i].p_hdr->eh_entries, -m);
err = ext4_ext_dirty(handle, inode, path + i);
if (err)
goto cleanup;
@@ -1369,7 +1367,7 @@ int ext4_ext_try_to_merge(struct inode *inode,
* sizeof(struct ext4_extent);
memmove(ex + 1, ex + 2, len);
}
- eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries) - 1);
+ le16_add_cpu(&eh->eh_entries, -1);
merge_done = 1;
WARN_ON(eh->eh_entries == 0);
if (!eh->eh_entries)
@@ -1560,7 +1558,7 @@ has_space:
path[depth].p_ext = nearex;
}
- eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)+1);
+ le16_add_cpu(&eh->eh_entries, 1);
nearex = path[depth].p_ext;
nearex->ee_block = newext->ee_block;
ext4_ext_store_pblock(nearex, ext_pblock(newext));
@@ -1699,7 +1697,7 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
err = ext4_ext_get_access(handle, inode, path);
if (err)
return err;
- path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1);
+ le16_add_cpu(&path->p_hdr->eh_entries, -1);
err = ext4_ext_dirty(handle, inode, path);
if (err)
return err;
@@ -1902,7 +1900,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
if (num == 0) {
/* this extent is removed; mark slot entirely unused */
ext4_ext_store_pblock(ex, 0);
- eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1);
+ le16_add_cpu(&eh->eh_entries, -1);
}
ex->ee_block = cpu_to_le32(block);
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index ac35ec58db55..20507a24506a 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -129,7 +129,7 @@ const struct file_operations ext4_file_operations = {
.write = do_sync_write,
.aio_read = generic_file_aio_read,
.aio_write = ext4_file_write,
- .ioctl = ext4_ioctl,
+ .unlocked_ioctl = ext4_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = ext4_compat_ioctl,
#endif
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index 8d50879d1c2c..a04a1ac4e0cf 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -72,6 +72,9 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync)
goto out;
}
+ if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
+ goto out;
+
/*
* The VFS has written the file data. If the inode is unaltered
* then we need not start a commit.
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 8036b9b5376b..5027af5e69ef 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -158,6 +158,7 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
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",
@@ -223,11 +224,9 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
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));
@@ -235,6 +234,12 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
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);
@@ -289,6 +294,80 @@ static int find_group_dir(struct super_block *sb, struct inode *parent,
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[best_flex].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 (best_flex < 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.
*
@@ -504,6 +583,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
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)
@@ -517,6 +597,12 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
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))
ret2 = find_group_dir(sb, dir, &group);
@@ -525,6 +611,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
} else
ret2 = find_group_other(sb, dir, &group);
+got_group:
err = -ENOSPC;
if (ret2 == -1)
goto out;
@@ -664,11 +751,9 @@ got:
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));
@@ -681,6 +766,13 @@ got:
percpu_counter_inc(&sbi->s_dirs_counter);
sb->s_dirt = 1;
+ 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;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 945cbf6cb1fc..0f86bbbb3263 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2501,12 +2501,10 @@ out_stop:
static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb,
unsigned long ino, struct ext4_iloc *iloc)
{
- unsigned long desc, group_desc;
ext4_group_t block_group;
unsigned long offset;
ext4_fsblk_t block;
- struct buffer_head *bh;
- struct ext4_group_desc * gdp;
+ struct ext4_group_desc *gdp;
if (!ext4_valid_inum(sb, ino)) {
/*
@@ -2518,22 +2516,10 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb,
}
block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
- if (block_group >= EXT4_SB(sb)->s_groups_count) {
- ext4_error(sb,"ext4_get_inode_block","group >= groups count");
+ gdp = ext4_get_group_desc(sb, block_group, NULL);
+ if (!gdp)
return 0;
- }
- smp_rmb();
- group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb);
- desc = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1);
- bh = EXT4_SB(sb)->s_group_desc[group_desc];
- if (!bh) {
- ext4_error (sb, "ext4_get_inode_block",
- "Descriptor not loaded");
- return 0;
- }
- gdp = (struct ext4_group_desc *)((__u8 *)bh->b_data +
- desc * EXT4_DESC_SIZE(sb));
/*
* Figure out the offset within the block group inode table
*/
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index 2ed7c37f897e..53a010bb0483 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -17,9 +17,9 @@
#include <linux/smp_lock.h>
#include <asm/uaccess.h>
-int ext4_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
- unsigned long arg)
+long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
+ struct inode *inode = filp->f_dentry->d_inode;
struct ext4_inode_info *ei = EXT4_I(inode);
unsigned int flags;
unsigned short rsv_window_size;
@@ -265,9 +265,6 @@ flags_err:
#ifdef CONFIG_COMPAT
long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
- struct inode *inode = file->f_path.dentry->d_inode;
- int ret;
-
/* These are just misnamed, they actually get/put from/to user an int */
switch (cmd) {
case EXT4_IOC32_GETFLAGS:
@@ -307,9 +304,6 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
default:
return -ENOIOCTLCMD;
}
- lock_kernel();
- ret = ext4_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg));
- unlock_kernel();
- return ret;
+ return ext4_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
}
#endif
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index ef97f19c2f9d..ad6bf37b92d5 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -576,8 +576,6 @@ static void ext4_mb_store_history(struct ext4_allocation_context *ac);
static struct proc_dir_entry *proc_root_ext4;
struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t);
-ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode,
- ext4_fsblk_t goal, unsigned long *count, int *errp);
static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
ext4_group_t group);
@@ -1168,8 +1166,9 @@ out:
return err;
}
-static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
- struct ext4_buddy *e4b)
+static noinline int
+ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
+ struct ext4_buddy *e4b)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
struct inode *inode = sbi->s_buddy_cache;
@@ -1965,7 +1964,8 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac,
return 0;
}
-static int ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
+static noinline int
+ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
{
ext4_group_t group;
ext4_group_t i;
@@ -2472,7 +2472,7 @@ static void ext4_mb_history_init(struct super_block *sb)
/* if we can't allocate history, then we simple won't use it */
}
-static void ext4_mb_store_history(struct ext4_allocation_context *ac)
+static noinline void ext4_mb_store_history(struct ext4_allocation_context *ac)
{
struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
struct ext4_mb_history h;
@@ -2808,7 +2808,7 @@ int ext4_mb_release(struct super_block *sb)
return 0;
}
-static void ext4_mb_free_committed_blocks(struct super_block *sb)
+static noinline void ext4_mb_free_committed_blocks(struct super_block *sb)
{
struct ext4_sb_info *sbi = EXT4_SB(sb);
int err;
@@ -3029,7 +3029,8 @@ void exit_ext4_mballoc(void)
* Check quota and mark choosed space (ac->ac_b_ex) non-free in bitmaps
* Returns 0 if success or error code
*/
-static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
+static noinline int
+ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
handle_t *handle)
{
struct buffer_head *bitmap_bh = NULL;
@@ -3102,9 +3103,7 @@ static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
ac->ac_b_ex.fe_group,
gdp));
}
- gdp->bg_free_blocks_count =
- cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)
- - ac->ac_b_ex.fe_len);
+ le16_add_cpu(&gdp->bg_free_blocks_count, -ac->ac_b_ex.fe_len);
gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp);
spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group));
percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len);
@@ -3146,15 +3145,16 @@ static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac)
* Normalization means making request better in terms of
* size and alignment
*/
-static void ext4_mb_normalize_request(struct ext4_allocation_context *ac,
+static noinline void
+ext4_mb_normalize_request(struct ext4_allocation_context *ac,
struct ext4_allocation_request *ar)
{
int bsbits, max;
ext4_lblk_t end;
- struct list_head *cur;
loff_t size, orig_size, start_off;
ext4_lblk_t start, orig_start;
struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
+ struct ext4_prealloc_space *pa;
/* do normalize only data requests, metadata requests
do not need preallocation */
@@ -3240,12 +3240,9 @@ static void ext4_mb_normalize_request(struct ext4_allocation_context *ac,
/* check we don't cross already preallocated blocks */
rcu_read_lock();
- list_for_each_rcu(cur, &ei->i_prealloc_list) {
- struct ext4_prealloc_space *pa;
+ list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) {
unsigned long pa_end;
- pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
-
if (pa->pa_deleted)
continue;
spin_lock(&pa->pa_lock);
@@ -3287,10 +3284,8 @@ static void ext4_mb_normalize_request(struct ext4_allocation_context *ac,
/* XXX: extra loop to check we really don't overlap preallocations */
rcu_read_lock();
- list_for_each_rcu(cur, &ei->i_prealloc_list) {
- struct ext4_prealloc_space *pa;
+ list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) {
unsigned long pa_end;
- pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
spin_lock(&pa->pa_lock);
if (pa->pa_deleted == 0) {
pa_end = pa->pa_lstart + pa->pa_len;
@@ -3412,12 +3407,11 @@ static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac,
/*
* search goal blocks in preallocated space
*/
-static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
+static noinline int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
{
struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
struct ext4_locality_group *lg;
struct ext4_prealloc_space *pa;
- struct list_head *cur;
/* only data can be preallocated */
if (!(ac->ac_flags & EXT4_MB_HINT_DATA))
@@ -3425,8 +3419,7 @@ static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
/* first, try per-file preallocation */
rcu_read_lock();
- list_for_each_rcu(cur, &ei->i_prealloc_list) {
- pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+ list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) {
/* all fields in this condition don't change,
* so we can skip locking for them */
@@ -3458,8 +3451,7 @@ static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
return 0;
rcu_read_lock();
- list_for_each_rcu(cur, &lg->lg_prealloc_list) {
- pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+ list_for_each_entry_rcu(pa, &lg->lg_prealloc_list, pa_inode_list) {
spin_lock(&pa->pa_lock);
if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) {
atomic_inc(&pa->pa_count);
@@ -3579,7 +3571,7 @@ static void ext4_mb_put_pa(struct ext4_allocation_context *ac,
/*
* creates new preallocated space for given inode
*/
-static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac)
+static noinline int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac)
{
struct super_block *sb = ac->ac_sb;
struct ext4_prealloc_space *pa;
@@ -3666,7 +3658,7 @@ static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac)
/*
* creates new preallocated space for locality group inodes belongs to
*/
-static int ext4_mb_new_group_pa(struct ext4_allocation_context *ac)
+static noinline int ext4_mb_new_group_pa(struct ext4_allocation_context *ac)
{
struct super_block *sb = ac->ac_sb;
struct ext4_locality_group *lg;
@@ -3739,7 +3731,7 @@ static int ext4_mb_new_preallocation(struct ext4_allocation_context *ac)
* the caller MUST hold group/inode locks.
* TODO: optimize the case when there are no in-core structures yet
*/
-static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
+static noinline int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
struct buffer_head *bitmap_bh,
struct ext4_prealloc_space *pa)
{
@@ -3811,7 +3803,7 @@ static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
return err;
}
-static int ext4_mb_release_group_pa(struct ext4_buddy *e4b,
+static noinline int ext4_mb_release_group_pa(struct ext4_buddy *e4b,
struct ext4_prealloc_space *pa)
{
struct ext4_allocation_context *ac;
@@ -3853,7 +3845,7 @@ static int ext4_mb_release_group_pa(struct ext4_buddy *e4b,
* - how many do we discard
* 1) how many requested
*/
-static int ext4_mb_discard_group_preallocations(struct super_block *sb,
+static noinline int ext4_mb_discard_group_preallocations(struct super_block *sb,
ext4_group_t group, int needed)
{
struct ext4_group_info *grp = ext4_get_group_info(sb, group);
@@ -4175,7 +4167,8 @@ static void ext4_mb_group_or_file(struct ext4_allocation_context *ac)
mutex_lock(&ac->ac_lg->lg_mutex);
}
-static int ext4_mb_initialize_context(struct ext4_allocation_context *ac,
+static noinline int
+ext4_mb_initialize_context(struct ext4_allocation_context *ac,
struct ext4_allocation_request *ar)
{
struct super_block *sb = ar->inode->i_sb;
@@ -4406,7 +4399,8 @@ static void ext4_mb_poll_new_transaction(struct super_block *sb,
ext4_mb_free_committed_blocks(sb);
}
-static int ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
+static noinline int
+ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
ext4_group_t group, ext4_grpblk_t block, int count)
{
struct ext4_group_info *db = e4b->bd_info;
@@ -4596,8 +4590,7 @@ do_more:
}
spin_lock(sb_bgl_lock(sbi, block_group));
- gdp->bg_free_blocks_count =
- cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + count);
+ le16_add_cpu(&gdp->bg_free_blocks_count, count);
gdp->bg_checksum = ext4_group_desc_csum(sbi, block_group, gdp);
spin_unlock(sb_bgl_lock(sbi, block_group));
percpu_counter_add(&sbi->s_freeblocks_counter, count);
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 28aa2ed4297e..da942bce6108 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -57,10 +57,15 @@ static struct buffer_head *ext4_append(handle_t *handle,
*block = inode->i_size >> inode->i_sb->s_blocksize_bits;
- if ((bh = ext4_bread(handle, inode, *block, 1, err))) {
+ bh = ext4_bread(handle, inode, *block, 1, err);
+ if (bh) {
inode->i_size += inode->i_sb->s_blocksize;
EXT4_I(inode)->i_disksize = inode->i_size;
- ext4_journal_get_write_access(handle,bh);
+ *err = ext4_journal_get_write_access(handle, bh);
+ if (*err) {
+ brelse(bh);
+ bh = NULL;
+ }
}
return bh;
}
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index e29efa0f9d62..728e3efa84b5 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -502,8 +502,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
EXT4_SB(sb)->s_gdb_count++;
kfree(o_group_desc);
- es->s_reserved_gdt_blocks =
- cpu_to_le16(le16_to_cpu(es->s_reserved_gdt_blocks) - 1);
+ le16_add_cpu(&es->s_reserved_gdt_blocks, -1);
ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh);
return 0;
@@ -877,8 +876,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
*/
ext4_blocks_count_set(es, ext4_blocks_count(es) +
input->blocks_count);
- es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) +
- EXT4_INODES_PER_GROUP(sb));
+ le32_add_cpu(&es->s_inodes_count, EXT4_INODES_PER_GROUP(sb));
/*
* We need to protect s_groups_count against other CPUs seeing
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 13383ba18f1d..cc00baf7cbd2 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -518,6 +518,7 @@ static void ext4_put_super (struct super_block * sb)
for (i = 0; i < sbi->s_gdb_count; i++)
brelse(sbi->s_group_desc[i]);
kfree(sbi->s_group_desc);
+ kfree(sbi->s_flex_groups);
percpu_counter_destroy(&sbi->s_freeblocks_counter);
percpu_counter_destroy(&sbi->s_freeinodes_counter);
percpu_counter_destroy(&sbi->s_dirs_counter);
@@ -1387,11 +1388,11 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
* a plain journaled filesystem we can keep it set as
* valid forever! :)
*/
- es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT4_VALID_FS);
+ es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
#endif
if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
- es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1);
+ le16_add_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);
@@ -1418,6 +1419,54 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
return res;
}
+static int ext4_fill_flex_info(struct super_block *sb)
+{
+ struct ext4_sb_info *sbi = EXT4_SB(sb);
+ struct ext4_group_desc *gdp = NULL;
+ struct buffer_head *bh;
+ ext4_group_t flex_group_count;
+ ext4_group_t flex_group;
+ int groups_per_flex = 0;
+ __u64 block_bitmap = 0;
+ int i;
+
+ if (!sbi->s_es->s_log_groups_per_flex) {
+ sbi->s_log_groups_per_flex = 0;
+ return 1;
+ }
+
+ sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
+ groups_per_flex = 1 << sbi->s_log_groups_per_flex;
+
+ flex_group_count = (sbi->s_groups_count + groups_per_flex - 1) /
+ groups_per_flex;
+ sbi->s_flex_groups = kmalloc(flex_group_count *
+ sizeof(struct flex_groups), GFP_KERNEL);
+ if (sbi->s_flex_groups == NULL) {
+ printk(KERN_ERR "EXT4-fs: not enough memory\n");
+ goto failed;
+ }
+ memset(sbi->s_flex_groups, 0, flex_group_count *
+ sizeof(struct flex_groups));
+
+ gdp = ext4_get_group_desc(sb, 1, &bh);
+ block_bitmap = ext4_block_bitmap(sb, gdp) - 1;
+
+ for (i = 0; i < sbi->s_groups_count; i++) {
+ gdp = ext4_get_group_desc(sb, i, &bh);
+
+ flex_group = ext4_flex_group(sbi, i);
+ sbi->s_flex_groups[flex_group].free_inodes +=
+ le16_to_cpu(gdp->bg_free_inodes_count);
+ sbi->s_flex_groups[flex_group].free_blocks +=
+ le16_to_cpu(gdp->bg_free_blocks_count);
+ }
+
+ return 1;
+failed:
+ return 0;
+}
+
__le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 block_group,
struct ext4_group_desc *gdp)
{
@@ -2115,6 +2164,14 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
printk(KERN_ERR "EXT4-fs: group descriptors corrupted!\n");
goto failed_mount2;
}
+ if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG))
+ if (!ext4_fill_flex_info(sb)) {
+ printk(KERN_ERR
+ "EXT4-fs: unable to initialize "
+ "flex_bg meta info!\n");
+ goto failed_mount2;
+ }
+
sbi->s_gdb_count = db_count;
get_random_bytes(&sbi->s_next_generation, sizeof(u32));
spin_lock_init(&sbi->s_next_gen_lock);
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index d7962139c010..b9a16d2ba9b6 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -484,8 +484,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode,
get_bh(bh);
ext4_forget(handle, 1, inode, bh, bh->b_blocknr);
} else {
- BHDR(bh)->h_refcount = cpu_to_le32(
- le32_to_cpu(BHDR(bh)->h_refcount) - 1);
+ le32_add_cpu(&BHDR(bh)->h_refcount, -1);
error = ext4_journal_dirty_metadata(handle, bh);
if (IS_SYNC(inode))
handle->h_sync = 1;
@@ -789,8 +788,7 @@ inserted:
if (error)
goto cleanup_dquot;
lock_buffer(new_bh);
- BHDR(new_bh)->h_refcount = cpu_to_le32(1 +
- le32_to_cpu(BHDR(new_bh)->h_refcount));
+ le32_add_cpu(&BHDR(new_bh)->h_refcount, 1);
ea_bdebug(new_bh, "reusing; refcount now=%d",
le32_to_cpu(BHDR(new_bh)->h_refcount));
unlock_buffer(new_bh);
@@ -808,10 +806,8 @@ inserted:
get_bh(new_bh);
} else {
/* We need to allocate a new block */
- ext4_fsblk_t goal = le32_to_cpu(
- EXT4_SB(sb)->s_es->s_first_data_block) +
- (ext4_fsblk_t)EXT4_I(inode)->i_block_group *
- EXT4_BLOCKS_PER_GROUP(sb);
+ ext4_fsblk_t goal = ext4_group_first_block_no(sb,
+ EXT4_I(inode)->i_block_group);
ext4_fsblk_t block = ext4_new_block(handle, inode,
goal, &error);
if (error)
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 96ba846992e9..67b6a989a0aa 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -219,7 +219,7 @@ static int jbd2_journal_start_thread(journal_t *journal)
if (IS_ERR(t))
return PTR_ERR(t);
- wait_event(journal->j_wait_done_commit, journal->j_task != 0);
+ wait_event(journal->j_wait_done_commit, journal->j_task != NULL);
return 0;
}
@@ -231,7 +231,8 @@ static void journal_kill_thread(journal_t *journal)
while (journal->j_task) {
wake_up(&journal->j_wait_commit);
spin_unlock(&journal->j_state_lock);
- wait_event(journal->j_wait_done_commit, journal->j_task == 0);
+ wait_event(journal->j_wait_done_commit,
+ journal->j_task == NULL);
spin_lock(&journal->j_state_lock);
}
spin_unlock(&journal->j_state_lock);
@@ -1969,14 +1970,14 @@ static int journal_init_jbd2_journal_head_cache(void)
{
int retval;
- J_ASSERT(jbd2_journal_head_cache == 0);
+ J_ASSERT(!jbd2_journal_head_cache);
jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head",
sizeof(struct journal_head),
0, /* offset */
SLAB_TEMPORARY, /* flags */
NULL); /* ctor */
retval = 0;
- if (jbd2_journal_head_cache == 0) {
+ if (!jbd2_journal_head_cache) {
retval = -ENOMEM;
printk(KERN_EMERG "JBD: no memory for journal_head cache\n");
}
@@ -2002,14 +2003,14 @@ static struct journal_head *journal_alloc_journal_head(void)
atomic_inc(&nr_journal_heads);
#endif
ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS);
- if (ret == 0) {
+ if (!ret) {
jbd_debug(1, "out of memory for journal_head\n");
if (time_after(jiffies, last_warning + 5*HZ)) {
printk(KERN_NOTICE "ENOMEM in %s, retrying.\n",
__FUNCTION__);
last_warning = jiffies;
}
- while (ret == 0) {
+ while (!ret) {
yield();
ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS);
}
diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c
index df36f42e19e1..2e1453a5e998 100644
--- a/fs/jbd2/revoke.c
+++ b/fs/jbd2/revoke.c
@@ -174,13 +174,13 @@ int __init jbd2_journal_init_revoke_caches(void)
0,
SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
NULL);
- if (jbd2_revoke_record_cache == 0)
+ if (!jbd2_revoke_record_cache)
return -ENOMEM;
jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table",
sizeof(struct jbd2_revoke_table_s),
0, SLAB_TEMPORARY, NULL);
- if (jbd2_revoke_table_cache == 0) {
+ if (!jbd2_revoke_table_cache) {
kmem_cache_destroy(jbd2_revoke_record_cache);
jbd2_revoke_record_cache = NULL;
return -ENOMEM;
diff --git a/include/linux/ext4_fs.h b/include/linux/ext4_fs.h
index 250032548597..22810b15ac8c 100644
--- a/include/linux/ext4_fs.h
+++ b/include/linux/ext4_fs.h
@@ -171,6 +171,15 @@ struct ext4_group_desc
__u32 bg_reserved2[3];
};
+/*
+ * Structure of a flex block group info
+ */
+
+struct flex_groups {
+ __u32 free_inodes;
+ __u32 free_blocks;
+};
+
#define EXT4_BG_INODE_UNINIT 0x0001 /* Inode table/bitmap not in use */
#define EXT4_BG_BLOCK_UNINIT 0x0002 /* Block bitmap not in use */
#define EXT4_BG_INODE_ZEROED 0x0004 /* On-disk itable initialized to zero */
@@ -648,7 +657,10 @@ struct ext4_super_block {
__le16 s_mmp_interval; /* # seconds to wait in MMP checking */
__le64 s_mmp_block; /* Block for multi-mount protection */
__le32 s_raid_stripe_width; /* blocks on all data disks (N*stride)*/
- __u32 s_reserved[163]; /* Padding to the end of the block */
+ __u8 s_log_groups_per_flex; /* FLEX_BG group size */
+ __u8 s_reserved_char_pad2;
+ __le16 s_reserved_pad;
+ __u32 s_reserved[162]; /* Padding to the end of the block */
};
#ifdef __KERNEL__
@@ -1049,8 +1061,7 @@ extern int ext4_block_truncate_page(handle_t *handle, struct page *page,
struct address_space *mapping, loff_t from);
/* ioctl.c */
-extern int ext4_ioctl (struct inode *, struct file *, unsigned int,
- unsigned long);
+extern long ext4_ioctl(struct file *, unsigned int, unsigned long);
extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long);
/* migrate.c */
@@ -1161,6 +1172,17 @@ struct ext4_group_info *ext4_get_group_info(struct super_block *sb,
}
+static inline ext4_group_t ext4_flex_group(struct ext4_sb_info *sbi,
+ ext4_group_t block_group)
+{
+ return block_group >> sbi->s_log_groups_per_flex;
+}
+
+static inline unsigned int ext4_flex_bg_size(struct ext4_sb_info *sbi)
+{
+ return 1 << sbi->s_log_groups_per_flex;
+}
+
#define ext4_std_error(sb, errno) \
do { \
if ((errno)) \
diff --git a/include/linux/ext4_fs_sb.h b/include/linux/ext4_fs_sb.h
index abaae2c8cccf..b419fa4b606a 100644
--- a/include/linux/ext4_fs_sb.h
+++ b/include/linux/ext4_fs_sb.h
@@ -143,6 +143,9 @@ struct ext4_sb_info {
/* locality groups */
struct ext4_locality_group *s_locality_groups;
+
+ unsigned int s_log_groups_per_flex;
+ struct flex_groups *s_flex_groups;
};
#endif /* _LINUX_EXT4_FS_SB */