Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6: (31 commits)
  dquot: Detect partial write error to quota file in write_blk() and add printk_ratelimit for quota error messages
  ocfs2: Fix lock inversion in quotas during umount
  ocfs2: Use __dquot_transfer to avoid lock inversion
  ocfs2: Fix NULL pointer deref when writing local dquot
  ocfs2: Fix estimate of credits needed for quota allocation
  ocfs2: Fix quota locking
  ocfs2: Avoid unnecessary block mapping when refreshing quota info
  ocfs2: Do not map blocks from local quota file on each write
  quota: Refactor dquot_transfer code so that OCFS2 can pass in its references
  quota: unify quota init condition in setattr
  quota: remove sb_has_quota_active in get/set_info
  quota: unify ->set_dqblk
  quota: unify ->get_dqblk
  ext3: make barrier options consistent with ext4
  quota: Make quota stat accounting lockless.
  suppress warning: "quotatypes" defined but not used
  ext3: Fix waiting on transaction during fsync
  jbd: Provide function to check whether transaction will issue data barrier
  ufs: add ufs speciffic ->setattr call
  BKL: Remove BKL from ext2 filesystem
  ...

Documentation/filesystems/ext3.txt

@@ -59,8 +59,19 @@ commit=nrsec	(*)	Ext3 can be told to sync all its data and metadata
 			Setting it to very large values will improve
 			performance.
 
-barrier=1		This enables/disables barriers.  barrier=0 disables
-			it, barrier=1 enables it.
+barrier=<0(*)|1>	This enables/disables the use of write barriers in
+barrier			the jbd code.  barrier=0 disables, barrier=1 enables.
+nobarrier	(*)	This also requires an IO stack which can support
+			barriers, and if jbd gets an error on a barrier
+			write, it will disable again with a warning.
+			Write barriers enforce proper on-disk ordering
+			of journal commits, making volatile disk write caches
+			safe to use, at some performance penalty.  If
+			your disks are battery-backed in one way or another,
+			disabling barriers may safely improve performance.
+			The mount options "barrier" and "nobarrier" can
+			also be used to enable or disable barriers, for
+			consistency with other ext3 mount options.
 
 orlov		(*)	This enables the new Orlov block allocator. It is
 			enabled by default.

fs/ext2/balloc.c

@@ -1331,6 +1331,12 @@ retry_alloc:
 			goto io_error;
 
 		free_blocks = le16_to_cpu(gdp->bg_free_blocks_count);
+		/*
+		 * skip this group (and avoid loading bitmap) if there
+		 * are no free blocks
+		 */
+		if (!free_blocks)
+			continue;
 		/*
 		 * skip this group if the number of
 		 * free blocks is less than half of the reservation

fs/ext2/ialloc.c

@@ -106,7 +106,7 @@ void ext2_free_inode (struct inode * inode)
 	struct super_block * sb = inode->i_sb;
 	int is_directory;
 	unsigned long ino;
-	struct buffer_head *bitmap_bh = NULL;
+	struct buffer_head *bitmap_bh;
 	unsigned long block_group;
 	unsigned long bit;
 	struct ext2_super_block * es;
@@ -135,14 +135,13 @@ void ext2_free_inode (struct inode * inode)
 	    ino > le32_to_cpu(es->s_inodes_count)) {
 		ext2_error (sb, "ext2_free_inode",
 			    "reserved or nonexistent inode %lu", ino);
-		goto error_return;
+		return;
 	}
 	block_group = (ino - 1) / EXT2_INODES_PER_GROUP(sb);
 	bit = (ino - 1) % EXT2_INODES_PER_GROUP(sb);
-	brelse(bitmap_bh);
 	bitmap_bh = read_inode_bitmap(sb, block_group);
 	if (!bitmap_bh)
-		goto error_return;
+		return;
 
 	/* Ok, now we can actually update the inode bitmaps.. */
 	if (!ext2_clear_bit_atomic(sb_bgl_lock(EXT2_SB(sb), block_group),
@@ -154,7 +153,7 @@ void ext2_free_inode (struct inode * inode)
 	mark_buffer_dirty(bitmap_bh);
 	if (sb->s_flags & MS_SYNCHRONOUS)
 		sync_dirty_buffer(bitmap_bh);
-error_return:
+
 	brelse(bitmap_bh);
 }
 

fs/ext2/inode.c

@@ -22,7 +22,6 @@
  *  Assorted race fixes, rewrite of ext2_get_block() by Al Viro, 2000
  */
 
-#include <linux/smp_lock.h>
 #include <linux/time.h>
 #include <linux/highuid.h>
 #include <linux/pagemap.h>
@@ -1406,11 +1405,11 @@ static int __ext2_write_inode(struct inode *inode, int do_sync)
 			       /* If this is the first large file
 				* created, add a flag to the superblock.
 				*/
-				lock_kernel();
+				spin_lock(&EXT2_SB(sb)->s_lock);
 				ext2_update_dynamic_rev(sb);
 				EXT2_SET_RO_COMPAT_FEATURE(sb,
 					EXT2_FEATURE_RO_COMPAT_LARGE_FILE);
-				unlock_kernel();
+				spin_unlock(&EXT2_SB(sb)->s_lock);
 				ext2_write_super(sb);
 			}
 		}
@@ -1467,7 +1466,7 @@ int ext2_setattr(struct dentry *dentry, struct iattr *iattr)
 	if (error)
 		return error;
 
-	if (iattr->ia_valid & ATTR_SIZE)
+	if (is_quota_modification(inode, iattr))
 		dquot_initialize(inode);
 	if ((iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid) ||
 	    (iattr->ia_valid & ATTR_GID && iattr->ia_gid != inode->i_gid)) {

fs/ext2/super.c

@@ -26,7 +26,6 @@
 #include <linux/random.h>
 #include <linux/buffer_head.h>
 #include <linux/exportfs.h>
-#include <linux/smp_lock.h>
 #include <linux/vfs.h>
 #include <linux/seq_file.h>
 #include <linux/mount.h>
@@ -39,7 +38,7 @@
 #include "xip.h"
 
 static void ext2_sync_super(struct super_block *sb,
-			    struct ext2_super_block *es);
+			    struct ext2_super_block *es, int wait);
 static int ext2_remount (struct super_block * sb, int * flags, char * data);
 static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf);
 static int ext2_sync_fs(struct super_block *sb, int wait);
@@ -52,9 +51,11 @@ void ext2_error (struct super_block * sb, const char * function,
 	struct ext2_super_block *es = sbi->s_es;
 
 	if (!(sb->s_flags & MS_RDONLY)) {
+		spin_lock(&sbi->s_lock);
 		sbi->s_mount_state |= EXT2_ERROR_FS;
 		es->s_state |= cpu_to_le16(EXT2_ERROR_FS);
-		ext2_sync_super(sb, es);
+		spin_unlock(&sbi->s_lock);
+		ext2_sync_super(sb, es, 1);
 	}
 
 	va_start(args, fmt);
@@ -84,6 +85,9 @@ void ext2_msg(struct super_block *sb, const char *prefix,
 	va_end(args);
 }
 
+/*
+ * This must be called with sbi->s_lock held.
+ */
 void ext2_update_dynamic_rev(struct super_block *sb)
 {
 	struct ext2_super_block *es = EXT2_SB(sb)->s_es;
@@ -115,8 +119,6 @@ static void ext2_put_super (struct super_block * sb)
 	int i;
 	struct ext2_sb_info *sbi = EXT2_SB(sb);
 
-	lock_kernel();
-
 	if (sb->s_dirt)
 		ext2_write_super(sb);
 
@@ -124,8 +126,10 @@ static void ext2_put_super (struct super_block * sb)
 	if (!(sb->s_flags & MS_RDONLY)) {
 		struct ext2_super_block *es = sbi->s_es;
 
+		spin_lock(&sbi->s_lock);
 		es->s_state = cpu_to_le16(sbi->s_mount_state);
-		ext2_sync_super(sb, es);
+		spin_unlock(&sbi->s_lock);
+		ext2_sync_super(sb, es, 1);
 	}
 	db_count = sbi->s_gdb_count;
 	for (i = 0; i < db_count; i++)
@@ -140,8 +144,6 @@ static void ext2_put_super (struct super_block * sb)
 	sb->s_fs_info = NULL;
 	kfree(sbi->s_blockgroup_lock);
 	kfree(sbi);
-
-	unlock_kernel();
 }
 
 static struct kmem_cache * ext2_inode_cachep;
@@ -209,6 +211,7 @@ static int ext2_show_options(struct seq_file *seq, struct vfsmount *vfs)
 	struct ext2_super_block *es = sbi->s_es;
 	unsigned long def_mount_opts;
 
+	spin_lock(&sbi->s_lock);
 	def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
 
 	if (sbi->s_sb_block != 1)
@@ -281,6 +284,7 @@ static int ext2_show_options(struct seq_file *seq, struct vfsmount *vfs)
 	if (!test_opt(sb, RESERVATION))
 		seq_puts(seq, ",noreservation");
 
+	spin_unlock(&sbi->s_lock);
 	return 0;
 }
 
@@ -606,7 +610,6 @@ static int ext2_setup_super (struct super_block * sb,
 	if (!le16_to_cpu(es->s_max_mnt_count))
 		es->s_max_mnt_count = cpu_to_le16(EXT2_DFL_MAX_MNT_COUNT);
 	le16_add_cpu(&es->s_mnt_count, 1);
-	ext2_write_super(sb);
 	if (test_opt (sb, DEBUG))
 		ext2_msg(sb, KERN_INFO, "%s, %s, bs=%lu, fs=%lu, gc=%lu, "
 			"bpg=%lu, ipg=%lu, mo=%04lx]",
@@ -767,6 +770,8 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_fs_info = sbi;
 	sbi->s_sb_block = sb_block;
 
+	spin_lock_init(&sbi->s_lock);
+
 	/*
 	 * See what the current blocksize for the device is, and
 	 * use that as the blocksize.  Otherwise (or if the blocksize
@@ -1079,7 +1084,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	if (EXT2_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL))
 		ext2_msg(sb, KERN_WARNING,
 			"warning: mounting ext3 filesystem as ext2");
-	ext2_setup_super (sb, es, sb->s_flags & MS_RDONLY);
+	if (ext2_setup_super (sb, es, sb->s_flags & MS_RDONLY))
+		sb->s_flags |= MS_RDONLY;
+	ext2_write_super(sb);
 	return 0;
 
 cantfind_ext2:
@@ -1120,30 +1127,26 @@ static void ext2_clear_super_error(struct super_block *sb)
 		 * be remapped.  Nothing we can do but to retry the
 		 * write and hope for the best.
 		 */
-		printk(KERN_ERR "EXT2-fs: %s previous I/O error to "
-		       "superblock detected", sb->s_id);
+		ext2_msg(sb, KERN_ERR,
+		       "previous I/O error to superblock detected\n");
 		clear_buffer_write_io_error(sbh);
 		set_buffer_uptodate(sbh);
 	}
 }
 
-static void ext2_commit_super (struct super_block * sb,
-			       struct ext2_super_block * es)
-{
-	ext2_clear_super_error(sb);
-	es->s_wtime = cpu_to_le32(get_seconds());
-	mark_buffer_dirty(EXT2_SB(sb)->s_sbh);
-	sb->s_dirt = 0;
-}
-
-static void ext2_sync_super(struct super_block *sb, struct ext2_super_block *es)
+static void ext2_sync_super(struct super_block *sb, struct ext2_super_block *es,
+			    int wait)
 {
 	ext2_clear_super_error(sb);
+	spin_lock(&EXT2_SB(sb)->s_lock);
 	es->s_free_blocks_count = cpu_to_le32(ext2_count_free_blocks(sb));
 	es->s_free_inodes_count = cpu_to_le32(ext2_count_free_inodes(sb));
 	es->s_wtime = cpu_to_le32(get_seconds());
+	/* unlock before we do IO */
+	spin_unlock(&EXT2_SB(sb)->s_lock);
 	mark_buffer_dirty(EXT2_SB(sb)->s_sbh);
-	sync_dirty_buffer(EXT2_SB(sb)->s_sbh);
+	if (wait)
+		sync_dirty_buffer(EXT2_SB(sb)->s_sbh);
 	sb->s_dirt = 0;
 }
 
@@ -1157,43 +1160,18 @@ static void ext2_sync_super(struct super_block *sb, struct ext2_super_block *es)
  * may have been checked while mounted and e2fsck may have
  * set s_state to EXT2_VALID_FS after some corrections.
  */
-
 static int ext2_sync_fs(struct super_block *sb, int wait)
 {
+	struct ext2_sb_info *sbi = EXT2_SB(sb);
 	struct ext2_super_block *es = EXT2_SB(sb)->s_es;
-	struct buffer_head *sbh = EXT2_SB(sb)->s_sbh;
-
-	lock_kernel();
-	if (buffer_write_io_error(sbh)) {
-		/*
-		 * Oh, dear.  A previous attempt to write the
-		 * superblock failed.  This could happen because the
-		 * USB device was yanked out.  Or it could happen to
-		 * be a transient write error and maybe the block will
-		 * be remapped.  Nothing we can do but to retry the
-		 * write and hope for the best.
-		 */
-		ext2_msg(sb, KERN_ERR,
-		       "previous I/O error to superblock detected\n");
-		clear_buffer_write_io_error(sbh);
-		set_buffer_uptodate(sbh);
-	}
 
+	spin_lock(&sbi->s_lock);
 	if (es->s_state & cpu_to_le16(EXT2_VALID_FS)) {
 		ext2_debug("setting valid to 0\n");
 		es->s_state &= cpu_to_le16(~EXT2_VALID_FS);
-		es->s_free_blocks_count =
-			cpu_to_le32(ext2_count_free_blocks(sb));
-		es->s_free_inodes_count =
-			cpu_to_le32(ext2_count_free_inodes(sb));
-		es->s_mtime = cpu_to_le32(get_seconds());
-		ext2_sync_super(sb, es);
-	} else {
-		ext2_commit_super(sb, es);
 	}
-	sb->s_dirt = 0;
-	unlock_kernel();
-
+	spin_unlock(&sbi->s_lock);
+	ext2_sync_super(sb, es, wait);
 	return 0;
 }
 
@@ -1215,7 +1193,7 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 	unsigned long old_sb_flags;
 	int err;
 
-	lock_kernel();
+	spin_lock(&sbi->s_lock);
 
 	/* Store the old options */
 	old_sb_flags = sb->s_flags;
@@ -1254,13 +1232,13 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 		sbi->s_mount_opt |= old_mount_opt & EXT2_MOUNT_XIP;
 	}
 	if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) {
-		unlock_kernel();
+		spin_unlock(&sbi->s_lock);
 		return 0;
 	}
 	if (*flags & MS_RDONLY) {
 		if (le16_to_cpu(es->s_state) & EXT2_VALID_FS ||
 		    !(sbi->s_mount_state & EXT2_VALID_FS)) {
-			unlock_kernel();
+			spin_unlock(&sbi->s_lock);
 			return 0;
 		}
 		/*
@@ -1269,6 +1247,8 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 		 */
 		es->s_state = cpu_to_le16(sbi->s_mount_state);
 		es->s_mtime = cpu_to_le32(get_seconds());
+		spin_unlock(&sbi->s_lock);
+		ext2_sync_super(sb, es, 1);
 	} else {
 		__le32 ret = EXT2_HAS_RO_COMPAT_FEATURE(sb,
 					       ~EXT2_FEATURE_RO_COMPAT_SUPP);
@@ -1288,16 +1268,16 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 		sbi->s_mount_state = le16_to_cpu(es->s_state);
 		if (!ext2_setup_super (sb, es, 0))
 			sb->s_flags &= ~MS_RDONLY;
+		spin_unlock(&sbi->s_lock);
+		ext2_write_super(sb);
 	}
-	ext2_sync_super(sb, es);
-	unlock_kernel();
 	return 0;
 restore_opts:
 	sbi->s_mount_opt = old_opts.s_mount_opt;
 	sbi->s_resuid = old_opts.s_resuid;
 	sbi->s_resgid = old_opts.s_resgid;
 	sb->s_flags = old_sb_flags;
-	unlock_kernel();
+	spin_unlock(&sbi->s_lock);
 	return err;
 }
 
@@ -1308,6 +1288,8 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf)
 	struct ext2_super_block *es = sbi->s_es;
 	u64 fsid;
 
+	spin_lock(&sbi->s_lock);
+
 	if (test_opt (sb, MINIX_DF))
 		sbi->s_overhead_last = 0;
 	else if (sbi->s_blocks_last != le32_to_cpu(es->s_blocks_count)) {
@@ -1362,6 +1344,7 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf)
 	       le64_to_cpup((void *)es->s_uuid + sizeof(u64));
 	buf->f_fsid.val[0] = fsid & 0xFFFFFFFFUL;
 	buf->f_fsid.val[1] = (fsid >> 32) & 0xFFFFFFFFUL;
+	spin_unlock(&sbi->s_lock);
 	return 0;
 }
 

fs/ext2/xattr.c

@@ -345,7 +345,9 @@ static void ext2_xattr_update_super_block(struct super_block *sb)
 	if (EXT2_HAS_COMPAT_FEATURE(sb, EXT2_FEATURE_COMPAT_EXT_ATTR))
 		return;
 
+	spin_lock(&EXT2_SB(sb)->s_lock);
 	EXT2_SET_COMPAT_FEATURE(sb, EXT2_FEATURE_COMPAT_EXT_ATTR);
+	spin_unlock(&EXT2_SB(sb)->s_lock);
 	sb->s_dirt = 1;
 	mark_buffer_dirty(EXT2_SB(sb)->s_sbh);
 }

fs/ext3/balloc.c

@@ -1583,6 +1583,12 @@ retry_alloc:
 		if (!gdp)
 			goto io_error;
 		free_blocks = le16_to_cpu(gdp->bg_free_blocks_count);
+		/*
+		 * skip this group (and avoid loading bitmap) if there
+		 * are no free blocks
+		 */
+		if (!free_blocks)
+			continue;
 		/*
 		 * skip this group if the number of
 		 * free blocks is less than half of the reservation

fs/ext3/fsync.c

@@ -48,7 +48,7 @@ int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync)
 	struct inode *inode = dentry->d_inode;
 	struct ext3_inode_info *ei = EXT3_I(inode);
 	journal_t *journal = EXT3_SB(inode->i_sb)->s_journal;
-	int ret = 0;
+	int ret, needs_barrier = 0;
 	tid_t commit_tid;
 
 	if (inode->i_sb->s_flags & MS_RDONLY)
@@ -70,28 +70,26 @@ int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync)
 	 *  (they were dirtied by commit).  But that's OK - the blocks are
 	 *  safe in-journal, which is all fsync() needs to ensure.
 	 */
-	if (ext3_should_journal_data(inode)) {
-		ret = ext3_force_commit(inode->i_sb);
-		goto out;
-	}
+	if (ext3_should_journal_data(inode))
+		return ext3_force_commit(inode->i_sb);
 
 	if (datasync)
 		commit_tid = atomic_read(&ei->i_datasync_tid);
 	else
 		commit_tid = atomic_read(&ei->i_sync_tid);
 
-	if (log_start_commit(journal, commit_tid)) {
-		log_wait_commit(journal, commit_tid);
-		goto out;
-	}
+	if (test_opt(inode->i_sb, BARRIER) &&
+	    !journal_trans_will_send_data_barrier(journal, commit_tid))
+		needs_barrier = 1;
+	log_start_commit(journal, commit_tid);
+	ret = log_wait_commit(journal, commit_tid);
 
 	/*
 	 * In case we didn't commit a transaction, we have to flush
 	 * disk caches manually so that data really is on persistent
 	 * storage
 	 */
-	if (test_opt(inode->i_sb, BARRIER))
+	if (needs_barrier)
 		blkdev_issue_flush(inode->i_sb->s_bdev, NULL);
-out:
 	return ret;
 }

fs/ext3/inode.c

@@ -3151,7 +3151,7 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr)
 	if (error)
 		return error;
 
-	if (ia_valid & ATTR_SIZE)
+	if (is_quota_modification(inode, attr))
 		dquot_initialize(inode);
 	if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
 		(ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) {

fs/ext3/super.c

@@ -653,8 +653,12 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_printf(seq, ",commit=%u",
 			   (unsigned) (sbi->s_commit_interval / HZ));
 	}
-	if (test_opt(sb, BARRIER))
-		seq_puts(seq, ",barrier=1");
+
+	/*
+	 * Always display barrier state so it's clear what the status is.
+	 */
+	seq_puts(seq, ",barrier=");
+	seq_puts(seq, test_opt(sb, BARRIER) ? "1" : "0");
 	if (test_opt(sb, NOBH))
 		seq_puts(seq, ",nobh");
 
@@ -810,8 +814,8 @@ enum {
 	Opt_data_err_abort, Opt_data_err_ignore,
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
 	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
-	Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize,
-	Opt_usrquota, Opt_grpquota
+	Opt_noquota, Opt_ignore, Opt_barrier, Opt_nobarrier, Opt_err,
+	Opt_resize, Opt_usrquota, Opt_grpquota
 };
 
 static const match_table_t tokens = {
@@ -865,6 +869,8 @@ static const match_table_t tokens = {
 	{Opt_quota, "quota"},
 	{Opt_usrquota, "usrquota"},
 	{Opt_barrier, "barrier=%u"},
+	{Opt_barrier, "barrier"},
+	{Opt_nobarrier, "nobarrier"},
 	{Opt_resize, "resize"},
 	{Opt_err, NULL},
 };
@@ -967,7 +973,11 @@ static int parse_options (char *options, struct super_block *sb,
 		int token;
 		if (!*p)
 			continue;
-
+		/*
+		 * Initialize args struct so we know whether arg was
+		 * found; some options take optional arguments.
+		 */
+		args[0].to = args[0].from = 0;
 		token = match_token(p, tokens, args);
 		switch (token) {
 		case Opt_bsd_df:
@@ -1215,9 +1225,15 @@ set_qf_format:
 		case Opt_abort:
 			set_opt(sbi->s_mount_opt, ABORT);
 			break;
+		case Opt_nobarrier:
+			clear_opt(sbi->s_mount_opt, BARRIER);
+			break;
 		case Opt_barrier:
-			if (match_int(&args[0], &option))
-				return 0;
+			if (args[0].from) {
+				if (match_int(&args[0], &option))
+					return 0;
+			} else
+				option = 1;	/* No argument, default to 1 */
 			if (option)
 				set_opt(sbi->s_mount_opt, BARRIER);
 			else
@@ -1890,21 +1906,6 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
 	spin_lock_init(&sbi->s_next_gen_lock);
 
-	err = percpu_counter_init(&sbi->s_freeblocks_counter,
-			ext3_count_free_blocks(sb));
-	if (!err) {
-		err = percpu_counter_init(&sbi->s_freeinodes_counter,
-				ext3_count_free_inodes(sb));
-	}
-	if (!err) {
-		err = percpu_counter_init(&sbi->s_dirs_counter,
-				ext3_count_dirs(sb));
-	}
-	if (err) {
-		ext3_msg(sb, KERN_ERR, "error: insufficient memory");
-		goto failed_mount3;
-	}
-
 	/* per fileystem reservation list head & lock */
 	spin_lock_init(&sbi->s_rsv_window_lock);
 	sbi->s_rsv_window_root = RB_ROOT;
@@ -1945,15 +1946,29 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	if (!test_opt(sb, NOLOAD) &&
 	    EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL)) {
 		if (ext3_load_journal(sb, es, journal_devnum))
-			goto failed_mount3;
+			goto failed_mount2;
 	} else if (journal_inum) {
 		if (ext3_create_journal(sb, es, journal_inum))
-			goto failed_mount3;
+			goto failed_mount2;
 	} else {
 		if (!silent)
 			ext3_msg(sb, KERN_ERR,
 				"error: no journal found. "
 				"mounting ext3 over ext2?");
+		goto failed_mount2;
+	}
+	err = percpu_counter_init(&sbi->s_freeblocks_counter,
+			ext3_count_free_blocks(sb));
+	if (!err) {
+		err = percpu_counter_init(&sbi->s_freeinodes_counter,
+				ext3_count_free_inodes(sb));
+	}
+	if (!err) {
+		err = percpu_counter_init(&sbi->s_dirs_counter,
+				ext3_count_dirs(sb));
+	}
+	if (err) {
+		ext3_msg(sb, KERN_ERR, "error: insufficient memory");
 		goto failed_mount3;
 	}
 
@@ -1978,7 +1993,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 			ext3_msg(sb, KERN_ERR,
 				"error: journal does not support "
 				"requested data journaling mode");
-			goto failed_mount4;
+			goto failed_mount3;
 		}
 	default:
 		break;
@@ -2001,19 +2016,19 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	if (IS_ERR(root)) {
 		ext3_msg(sb, KERN_ERR, "error: get root inode failed");
 		ret = PTR_ERR(root);
-		goto failed_mount4;
+		goto failed_mount3;
 	}
 	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
 		iput(root);
 		ext3_msg(sb, KERN_ERR, "error: corrupt root inode, run e2fsck");
-		goto failed_mount4;
+		goto failed_mount3;
 	}
 	sb->s_root = d_alloc_root(root);
 	if (!sb->s_root) {
 		ext3_msg(sb, KERN_ERR, "error: get root dentry failed");
 		iput(root);
 		ret = -ENOMEM;
-		goto failed_mount4;
+		goto failed_mount3;
 	}
 
 	ext3_setup_super (sb, es, sb->s_flags & MS_RDONLY);
@@ -2039,12 +2054,11 @@ cantfind_ext3:
 		       sb->s_id);
 	goto failed_mount;
 
-failed_mount4:
-	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);
+	journal_destroy(sbi->s_journal);
 failed_mount2:
 	for (i = 0; i < db_count; i++)
 		brelse(sbi->s_group_desc[i]);
@@ -2278,6 +2292,9 @@ static int ext3_load_journal(struct super_block *sb,
 			return -EINVAL;
 	}
 
+	if (!(journal->j_flags & JFS_BARRIER))
+		printk(KERN_INFO "EXT3-fs: barriers not enabled\n");
+
 	if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) {
 		err = journal_update_format(journal);
 		if (err)  {

fs/ext4/inode.c

@@ -5425,7 +5425,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 	if (error)
 		return error;
 
-	if (ia_valid & ATTR_SIZE)
+	if (is_quota_modification(inode, attr))
 		dquot_initialize(inode);
 	if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
 		(ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) {

fs/gfs2/quota.c

@@ -1476,8 +1476,8 @@ static int gfs2_quota_get_xstate(struct super_block *sb,
 	return 0;
 }
 
-static int gfs2_xquota_get(struct super_block *sb, int type, qid_t id,
-			   struct fs_disk_quota *fdq)
+static int gfs2_get_dqblk(struct super_block *sb, int type, qid_t id,
+			  struct fs_disk_quota *fdq)
 {
 	struct gfs2_sbd *sdp = sb->s_fs_info;
 	struct gfs2_quota_lvb *qlvb;
@@ -1521,8 +1521,8 @@ out:
 /* GFS2 only supports a subset of the XFS fields */
 #define GFS2_FIELDMASK (FS_DQ_BSOFT|FS_DQ_BHARD)
 
-static int gfs2_xquota_set(struct super_block *sb, int type, qid_t id,
-			   struct fs_disk_quota *fdq)
+static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id,
+			  struct fs_disk_quota *fdq)
 {
 	struct gfs2_sbd *sdp = sb->s_fs_info;
 	struct gfs2_inode *ip = GFS2_I(sdp->sd_quota_inode);
@@ -1629,7 +1629,7 @@ out_put:
 const struct quotactl_ops gfs2_quotactl_ops = {
 	.quota_sync     = gfs2_quota_sync,
 	.get_xstate     = gfs2_quota_get_xstate,
-	.get_xquota	= gfs2_xquota_get,
-	.set_xquota	= gfs2_xquota_set,
+	.get_dqblk	= gfs2_get_dqblk,
+	.set_dqblk	= gfs2_set_dqblk,
 };
 

fs/jbd/commit.c

@@ -786,6 +786,12 @@ wait_for_iobuf:
 
 	jbd_debug(3, "JBD: commit phase 6\n");
 
+	/* All metadata is written, now write commit record and do cleanup */
+	spin_lock(&journal->j_state_lock);
+	J_ASSERT(commit_transaction->t_state == T_COMMIT);
+	commit_transaction->t_state = T_COMMIT_RECORD;
+	spin_unlock(&journal->j_state_lock);
+
 	if (journal_write_commit_record(journal, commit_transaction))
 		err = -EIO;
 
@@ -923,7 +929,7 @@ restart_loop:
 
 	jbd_debug(3, "JBD: commit phase 8\n");
 
-	J_ASSERT(commit_transaction->t_state == T_COMMIT);
+	J_ASSERT(commit_transaction->t_state == T_COMMIT_RECORD);
 
 	commit_transaction->t_state = T_FINISHED;
 	J_ASSERT(commit_transaction == journal->j_committing_transaction);

fs/jbd/journal.c

@@ -564,6 +564,38 @@ int log_wait_commit(journal_t *journal, tid_t tid)
 	return err;
 }
 
+/*
+ * Return 1 if a given transaction has not yet sent barrier request
+ * connected with a transaction commit. If 0 is returned, transaction
+ * may or may not have sent the barrier. Used to avoid sending barrier
+ * twice in common cases.
+ */
+int journal_trans_will_send_data_barrier(journal_t *journal, tid_t tid)
+{
+	int ret = 0;
+	transaction_t *commit_trans;
+
+	if (!(journal->j_flags & JFS_BARRIER))
+		return 0;
+	spin_lock(&journal->j_state_lock);
+	/* Transaction already committed? */
+	if (tid_geq(journal->j_commit_sequence, tid))
+		goto out;
+	/*
+	 * Transaction is being committed and we already proceeded to
+	 * writing commit record?
+	 */
+	commit_trans = journal->j_committing_transaction;
+	if (commit_trans && commit_trans->t_tid == tid &&
+	    commit_trans->t_state >= T_COMMIT_RECORD)
+		goto out;
+	ret = 1;
+out:
+	spin_unlock(&journal->j_state_lock);
+	return ret;
+}
+EXPORT_SYMBOL(journal_trans_will_send_data_barrier);
+
 /*
  * Log buffer allocation routines:
  */
@@ -1157,6 +1189,7 @@ int journal_destroy(journal_t *journal)
 {
 	int err = 0;
 
+	
 	/* Wait for the commit thread to wake up and die. */
 	journal_kill_thread(journal);
 

fs/jfs/file.c

@@ -98,7 +98,7 @@ int jfs_setattr(struct dentry *dentry, struct iattr *iattr)
 	if (rc)
 		return rc;
 
-	if (iattr->ia_valid & ATTR_SIZE)
+	if (is_quota_modification(inode, iattr))
 		dquot_initialize(inode);
 	if ((iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid) ||
 	    (iattr->ia_valid & ATTR_GID && iattr->ia_gid != inode->i_gid)) {

fs/ocfs2/dlmglue.c

@@ -3897,7 +3897,8 @@ static int ocfs2_refresh_qinfo(struct ocfs2_mem_dqinfo *oinfo)
 		oinfo->dqi_gi.dqi_free_entry =
 					be32_to_cpu(lvb->lvb_free_entry);
 	} else {
-		status = ocfs2_read_quota_block(oinfo->dqi_gqinode, 0, &bh);
+		status = ocfs2_read_quota_phys_block(oinfo->dqi_gqinode,
+						     oinfo->dqi_giblk, &bh);
 		if (status) {
 			mlog_errno(status);
 			goto bail;

fs/ocfs2/file.c

@@ -933,9 +933,8 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
 	struct ocfs2_super *osb = OCFS2_SB(sb);
 	struct buffer_head *bh = NULL;
 	handle_t *handle = NULL;
-	int qtype;
-	struct dquot *transfer_from[MAXQUOTAS] = { };
 	struct dquot *transfer_to[MAXQUOTAS] = { };
+	int qtype;
 
 	mlog_entry("(0x%p, '%.*s')\n", dentry,
 	           dentry->d_name.len, dentry->d_name.name);
@@ -966,10 +965,10 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
 	if (status)
 		return status;
 
+	if (is_quota_modification(inode, attr))
+		dquot_initialize(inode);
 	size_change = S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE;
 	if (size_change) {
-		dquot_initialize(inode);
-
 		status = ocfs2_rw_lock(inode, 1);
 		if (status < 0) {
 			mlog_errno(status);
@@ -1019,9 +1018,7 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
 		    OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
 			transfer_to[USRQUOTA] = dqget(sb, attr->ia_uid,
 						      USRQUOTA);
-			transfer_from[USRQUOTA] = dqget(sb, inode->i_uid,
-							USRQUOTA);
-			if (!transfer_to[USRQUOTA] || !transfer_from[USRQUOTA]) {
+			if (!transfer_to[USRQUOTA]) {
 				status = -ESRCH;
 				goto bail_unlock;
 			}
@@ -1031,9 +1028,7 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
 		    OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
 			transfer_to[GRPQUOTA] = dqget(sb, attr->ia_gid,
 						      GRPQUOTA);
-			transfer_from[GRPQUOTA] = dqget(sb, inode->i_gid,
-							GRPQUOTA);
-			if (!transfer_to[GRPQUOTA] || !transfer_from[GRPQUOTA]) {
+			if (!transfer_to[GRPQUOTA]) {
 				status = -ESRCH;
 				goto bail_unlock;
 			}
@@ -1045,7 +1040,7 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
 			mlog_errno(status);
 			goto bail_unlock;
 		}
-		status = dquot_transfer(inode, attr);
+		status = __dquot_transfer(inode, transfer_to);
 		if (status < 0)
 			goto bail_commit;
 	} else {
@@ -1085,10 +1080,8 @@ bail:
 	brelse(bh);
 
 	/* Release quota pointers in case we acquired them */
-	for (qtype = 0; qtype < MAXQUOTAS; qtype++) {
+	for (qtype = 0; qtype < MAXQUOTAS; qtype++)
 		dqput(transfer_to[qtype]);
-		dqput(transfer_from[qtype]);
-	}
 
 	if (!status && attr->ia_valid & ATTR_MODE) {
 		status = ocfs2_acl_chmod(inode);

fs/ocfs2/quota.h

@@ -23,6 +23,7 @@
 struct ocfs2_dquot {
 	struct dquot dq_dquot;	/* Generic VFS dquot */
 	loff_t dq_local_off;	/* Offset in the local quota file */
+	u64 dq_local_phys_blk;	/* Physical block carrying quota structure */
 	struct ocfs2_quota_chunk *dq_chunk;	/* Chunk dquot is in */
 	unsigned int dq_use_count;	/* Number of nodes having reference to this entry in global quota file */
 	s64 dq_origspace;	/* Last globally synced space usage */
@@ -51,8 +52,9 @@ struct ocfs2_mem_dqinfo {
 	struct ocfs2_lock_res dqi_gqlock;	/* Lock protecting quota information structure */
 	struct buffer_head *dqi_gqi_bh;	/* Buffer head with global quota file inode - set only if inode lock is obtained */
 	int dqi_gqi_count;		/* Number of holders of dqi_gqi_bh */
+	u64 dqi_giblk;			/* Number of block with global information header */
 	struct buffer_head *dqi_lqi_bh;	/* Buffer head with local quota file inode */
-	struct buffer_head *dqi_ibh;	/* Buffer with information header */
+	struct buffer_head *dqi_libh;	/* Buffer with local information header */
 	struct qtree_mem_dqinfo dqi_gi;	/* Info about global file */
 	struct delayed_work dqi_sync_work;	/* Work for syncing dquots */
 	struct ocfs2_quota_recovery *dqi_rec;	/* Pointer to recovery
@@ -102,8 +104,12 @@ static inline int ocfs2_global_release_dquot(struct dquot *dquot)
 
 int ocfs2_lock_global_qf(struct ocfs2_mem_dqinfo *oinfo, int ex);
 void ocfs2_unlock_global_qf(struct ocfs2_mem_dqinfo *oinfo, int ex);
-int ocfs2_read_quota_block(struct inode *inode, u64 v_block,
-			   struct buffer_head **bh);
+int ocfs2_validate_quota_block(struct super_block *sb, struct buffer_head *bh);
+int ocfs2_read_quota_phys_block(struct inode *inode, u64 p_block,
+				struct buffer_head **bh);
+int ocfs2_create_local_dquot(struct dquot *dquot);
+int ocfs2_local_release_dquot(handle_t *handle, struct dquot *dquot);
+int ocfs2_local_write_dquot(struct dquot *dquot);
 
 extern const struct dquot_operations ocfs2_quota_operations;
 extern struct quota_format_type ocfs2_quota_format;

fs/ocfs2/quota_global.c

@@ -25,8 +25,44 @@
 #include "dlmglue.h"
 #include "uptodate.h"
 #include "super.h"
+#include "buffer_head_io.h"
 #include "quota.h"
 
+/*
+ * Locking of quotas with OCFS2 is rather complex. Here are rules that
+ * should be obeyed by all the functions:
+ * - any write of quota structure (either to local or global file) is protected
+ *   by dqio_mutex or dquot->dq_lock.
+ * - any modification of global quota file holds inode cluster lock, i_mutex,
+ *   and ip_alloc_sem of the global quota file (achieved by
+ *   ocfs2_lock_global_qf). It also has to hold qinfo_lock.
+ * - an allocation of new blocks for local quota file is protected by
+ *   its ip_alloc_sem
+ *
+ * A rough sketch of locking dependencies (lf = local file, gf = global file):
+ * Normal filesystem operation:
+ *   start_trans -> dqio_mutex -> write to lf
+ * Syncing of local and global file:
+ *   ocfs2_lock_global_qf -> start_trans -> dqio_mutex -> qinfo_lock ->
+ *     write to gf
+ *						       -> write to lf
+ * Acquire dquot for the first time:
+ *   dq_lock -> ocfs2_lock_global_qf -> qinfo_lock -> read from gf
+ *				     -> alloc space for gf
+ *				     -> start_trans -> qinfo_lock -> write to gf
+ *	     -> ip_alloc_sem of lf -> alloc space for lf
+ *	     -> write to lf
+ * Release last reference to dquot:
+ *   dq_lock -> ocfs2_lock_global_qf -> start_trans -> qinfo_lock -> write to gf
+ *	     -> write to lf
+ * Note that all the above operations also hold the inode cluster lock of lf.
+ * Recovery:
+ *   inode cluster lock of recovered lf
+ *     -> read bitmaps -> ip_alloc_sem of lf
+ *     -> ocfs2_lock_global_qf -> start_trans -> dqio_mutex -> qinfo_lock ->
+ *        write to gf
+ */
+
 static struct workqueue_struct *ocfs2_quota_wq = NULL;
 
 static void qsync_work_fn(struct work_struct *work);
@@ -91,8 +127,7 @@ struct qtree_fmt_operations ocfs2_global_ops = {
 	.is_id = ocfs2_global_is_id,
 };
 
-static int ocfs2_validate_quota_block(struct super_block *sb,
-				      struct buffer_head *bh)
+int ocfs2_validate_quota_block(struct super_block *sb, struct buffer_head *bh)
 {
 	struct ocfs2_disk_dqtrailer *dqt =
 		ocfs2_block_dqtrailer(sb->s_blocksize, bh->b_data);
@@ -110,54 +145,19 @@ static int ocfs2_validate_quota_block(struct super_block *sb,
 	return ocfs2_validate_meta_ecc(sb, bh->b_data, &dqt->dq_check);
 }
 
-int ocfs2_read_quota_block(struct inode *inode, u64 v_block,
-			   struct buffer_head **bh)
+int ocfs2_read_quota_phys_block(struct inode *inode, u64 p_block,
+				struct buffer_head **bhp)
 {
-	int rc = 0;
-	struct buffer_head *tmp = *bh;
-
-	if (i_size_read(inode) >> inode->i_sb->s_blocksize_bits <= v_block) {
-		ocfs2_error(inode->i_sb,
-			    "Quota file %llu is probably corrupted! Requested "
-			    "to read block %Lu but file has size only %Lu\n",
-			    (unsigned long long)OCFS2_I(inode)->ip_blkno,
-			    (unsigned long long)v_block,
-			    (unsigned long long)i_size_read(inode));
-		return -EIO;
-	}
-	rc = ocfs2_read_virt_blocks(inode, v_block, 1, &tmp, 0,
-				    ocfs2_validate_quota_block);
+	int rc;
+
+	*bhp = NULL;
+	rc = ocfs2_read_blocks(INODE_CACHE(inode), p_block, 1, bhp, 0,
+			       ocfs2_validate_quota_block);
 	if (rc)
 		mlog_errno(rc);
-
-	/* If ocfs2_read_virt_blocks() got us a new bh, pass it up. */
-	if (!rc && !*bh)
-		*bh = tmp;
-
 	return rc;
 }
 
-static int ocfs2_get_quota_block(struct inode *inode, int block,
-				 struct buffer_head **bh)
-{
-	u64 pblock, pcount;
-	int err;
-
-	down_read(&OCFS2_I(inode)->ip_alloc_sem);
-	err = ocfs2_extent_map_get_blocks(inode, block, &pblock, &pcount, NULL);
-	up_read(&OCFS2_I(inode)->ip_alloc_sem);
-	if (err) {
-		mlog_errno(err);
-		return err;
-	}
-	*bh = sb_getblk(inode->i_sb, pblock);
-	if (!*bh) {
-		err = -EIO;
-		mlog_errno(err);
-	}
-	return err;
-}
-
 /* Read data from global quotafile - avoid pagecache and such because we cannot
  * afford acquiring the locks... We use quota cluster lock to serialize
  * operations. Caller is responsible for acquiring it. */
@@ -172,6 +172,7 @@ ssize_t ocfs2_quota_read(struct super_block *sb, int type, char *data,
 	int err = 0;
 	struct buffer_head *bh;
 	size_t toread, tocopy;
+	u64 pblock = 0, pcount = 0;
 
 	if (off > i_size)
 		return 0;
@@ -180,8 +181,19 @@ ssize_t ocfs2_quota_read(struct super_block *sb, int type, char *data,
 	toread = len;
 	while (toread > 0) {
 		tocopy = min_t(size_t, (sb->s_blocksize - offset), toread);
+		if (!pcount) {
+			err = ocfs2_extent_map_get_blocks(gqinode, blk, &pblock,
+							  &pcount, NULL);
+			if (err) {
+				mlog_errno(err);
+				return err;
+			}
+		} else {
+			pcount--;
+			pblock++;
+		}
 		bh = NULL;
-		err = ocfs2_read_quota_block(gqinode, blk, &bh);
+		err = ocfs2_read_quota_phys_block(gqinode, pblock, &bh);
 		if (err) {
 			mlog_errno(err);
 			return err;
@@ -209,6 +221,7 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
 	int err = 0, new = 0, ja_type;
 	struct buffer_head *bh = NULL;
 	handle_t *handle = journal_current_handle();
+	u64 pblock, pcount;
 
 	if (!handle) {
 		mlog(ML_ERROR, "Quota write (off=%llu, len=%llu) cancelled "
@@ -221,12 +234,11 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
 		len = sb->s_blocksize - OCFS2_QBLK_RESERVED_SPACE - offset;
 	}
 
-	mutex_lock_nested(&gqinode->i_mutex, I_MUTEX_QUOTA);
 	if (gqinode->i_size < off + len) {
 		loff_t rounded_end =
 				ocfs2_align_bytes_to_blocks(sb, off + len);
 
-		/* Space is already allocated in ocfs2_global_read_dquot() */
+		/* Space is already allocated in ocfs2_acquire_dquot() */
 		err = ocfs2_simple_size_update(gqinode,
 					       oinfo->dqi_gqi_bh,
 					       rounded_end);
@@ -234,13 +246,20 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
 			goto out;
 		new = 1;
 	}
+	err = ocfs2_extent_map_get_blocks(gqinode, blk, &pblock, &pcount, NULL);
+	if (err) {
+		mlog_errno(err);
+		goto out;
+	}
 	/* Not rewriting whole block? */
 	if ((offset || len < sb->s_blocksize - OCFS2_QBLK_RESERVED_SPACE) &&
 	    !new) {
-		err = ocfs2_read_quota_block(gqinode, blk, &bh);
+		err = ocfs2_read_quota_phys_block(gqinode, pblock, &bh);
 		ja_type = OCFS2_JOURNAL_ACCESS_WRITE;
 	} else {
-		err = ocfs2_get_quota_block(gqinode, blk, &bh);
+		bh = sb_getblk(sb, pblock);
+		if (!bh)
+			err = -ENOMEM;
 		ja_type = OCFS2_JOURNAL_ACCESS_CREATE;
 	}
 	if (err) {
@@ -265,13 +284,11 @@ ssize_t ocfs2_quota_write(struct super_block *sb, int type,
 	brelse(bh);
 out:
 	if (err) {
-		mutex_unlock(&gqinode->i_mutex);
 		mlog_errno(err);
 		return err;
 	}
 	gqinode->i_version++;
 	ocfs2_mark_inode_dirty(handle, gqinode, oinfo->dqi_gqi_bh);
-	mutex_unlock(&gqinode->i_mutex);
 	return len;
 }
 
@@ -289,11 +306,23 @@ int ocfs2_lock_global_qf(struct ocfs2_mem_dqinfo *oinfo, int ex)
 	else
 		WARN_ON(bh != oinfo->dqi_gqi_bh);
 	spin_unlock(&dq_data_lock);
+	if (ex) {
+		mutex_lock(&oinfo->dqi_gqinode->i_mutex);
+		down_write(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
+	} else {
+		down_read(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
+	}
 	return 0;
 }
 
 void ocfs2_unlock_global_qf(struct ocfs2_mem_dqinfo *oinfo, int ex)
 {
+	if (ex) {
+		up_write(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
+		mutex_unlock(&oinfo->dqi_gqinode->i_mutex);
+	} else {
+		up_read(&OCFS2_I(oinfo->dqi_gqinode)->ip_alloc_sem);
+	}
 	ocfs2_inode_unlock(oinfo->dqi_gqinode, ex);
 	brelse(oinfo->dqi_gqi_bh);
 	spin_lock(&dq_data_lock);
@@ -311,6 +340,7 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
 	struct ocfs2_global_disk_dqinfo dinfo;
 	struct mem_dqinfo *info = sb_dqinfo(sb, type);
 	struct ocfs2_mem_dqinfo *oinfo = info->dqi_priv;
+	u64 pcount;
 	int status;
 
 	mlog_entry_void();
@@ -337,9 +367,19 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
 		mlog_errno(status);
 		goto out_err;
 	}
+
+	status = ocfs2_extent_map_get_blocks(gqinode, 0, &oinfo->dqi_giblk,
+					     &pcount, NULL);
+	if (status < 0)
+		goto out_unlock;
+
+	status = ocfs2_qinfo_lock(oinfo, 0);
+	if (status < 0)
+		goto out_unlock;
 	status = sb->s_op->quota_read(sb, type, (char *)&dinfo,
 				      sizeof(struct ocfs2_global_disk_dqinfo),
 				      OCFS2_GLOBAL_INFO_OFF);
+	ocfs2_qinfo_unlock(oinfo, 0);
 	ocfs2_unlock_global_qf(oinfo, 0);
 	if (status != sizeof(struct ocfs2_global_disk_dqinfo)) {
 		mlog(ML_ERROR, "Cannot read global quota info (%d).\n",
@@ -366,6 +406,10 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
 out_err:
 	mlog_exit(status);
 	return status;
+out_unlock:
+	ocfs2_unlock_global_qf(oinfo, 0);
+	mlog_errno(status);
+	goto out_err;
 }
 
 /* Write information to global quota file. Expects exlusive lock on quota
@@ -424,78 +468,10 @@ static int ocfs2_global_qinit_alloc(struct super_block *sb, int type)
 
 static int ocfs2_calc_global_qinit_credits(struct super_block *sb, int type)
 {
-	/* We modify all the allocated blocks, tree root, and info block */
+	/* We modify all the allocated blocks, tree root, info block and
+	 * the inode */
 	return (ocfs2_global_qinit_alloc(sb, type) + 2) *
-			OCFS2_QUOTA_BLOCK_UPDATE_CREDITS;
-}
-
-/* Read in information from global quota file and acquire a reference to it.
- * dquot_acquire() has already started the transaction and locked quota file */
-int ocfs2_global_read_dquot(struct dquot *dquot)
-{
-	int err, err2, ex = 0;
-	struct super_block *sb = dquot->dq_sb;
-	int type = dquot->dq_type;
-	struct ocfs2_mem_dqinfo *info = sb_dqinfo(sb, type)->dqi_priv;
-	struct ocfs2_super *osb = OCFS2_SB(sb);
-	struct inode *gqinode = info->dqi_gqinode;
-	int need_alloc = ocfs2_global_qinit_alloc(sb, type);
-	handle_t *handle = NULL;
-
-	err = ocfs2_qinfo_lock(info, 0);
-	if (err < 0)
-		goto out;
-	err = qtree_read_dquot(&info->dqi_gi, dquot);
-	if (err < 0)
-		goto out_qlock;
-	OCFS2_DQUOT(dquot)->dq_use_count++;
-	OCFS2_DQUOT(dquot)->dq_origspace = dquot->dq_dqb.dqb_curspace;
-	OCFS2_DQUOT(dquot)->dq_originodes = dquot->dq_dqb.dqb_curinodes;
-	ocfs2_qinfo_unlock(info, 0);
-
-	if (!dquot->dq_off) {	/* No real quota entry? */
-		ex = 1;
-		/*
-		 * Add blocks to quota file before we start a transaction since
-		 * locking allocators ranks above a transaction start
-		 */
-		WARN_ON(journal_current_handle());
-		down_write(&OCFS2_I(gqinode)->ip_alloc_sem);
-		err = ocfs2_extend_no_holes(gqinode,
-			gqinode->i_size + (need_alloc << sb->s_blocksize_bits),
-			gqinode->i_size);
-		up_write(&OCFS2_I(gqinode)->ip_alloc_sem);
-		if (err < 0)
-			goto out;
-	}
-
-	handle = ocfs2_start_trans(osb,
-				   ocfs2_calc_global_qinit_credits(sb, type));
-	if (IS_ERR(handle)) {
-		err = PTR_ERR(handle);
-		goto out;
-	}
-	err = ocfs2_qinfo_lock(info, ex);
-	if (err < 0)
-		goto out_trans;
-	err = qtree_write_dquot(&info->dqi_gi, dquot);
-	if (ex && info_dirty(sb_dqinfo(dquot->dq_sb, dquot->dq_type))) {
-		err2 = __ocfs2_global_write_info(dquot->dq_sb, dquot->dq_type);
-		if (!err)
-			err = err2;
-	}
-out_qlock:
-	if (ex)
-		ocfs2_qinfo_unlock(info, 1);
-	else
-		ocfs2_qinfo_unlock(info, 0);
-out_trans:
-	if (handle)
-		ocfs2_commit_trans(osb, handle);
-out:
-	if (err < 0)
-		mlog_errno(err);
-	return err;
+			OCFS2_QUOTA_BLOCK_UPDATE_CREDITS + 1;
 }
 
 /* Sync local information about quota modifications with global quota file.
@@ -636,14 +612,13 @@ static int ocfs2_sync_dquot_helper(struct dquot *dquot, unsigned long type)
 	}
 	mutex_lock(&sb_dqopt(sb)->dqio_mutex);
 	status = ocfs2_sync_dquot(dquot);
-	mutex_unlock(&sb_dqopt(sb)->dqio_mutex);
 	if (status < 0)
 		mlog_errno(status);
 	/* We have to write local structure as well... */
-	dquot_mark_dquot_dirty(dquot);
-	status = dquot_commit(dquot);
+	status = ocfs2_local_write_dquot(dquot);
 	if (status < 0)
 		mlog_errno(status);
+	mutex_unlock(&sb_dqopt(sb)->dqio_mutex);
 	ocfs2_commit_trans(osb, handle);
 out_ilock:
 	ocfs2_unlock_global_qf(oinfo, 1);
@@ -682,7 +657,9 @@ static int ocfs2_write_dquot(struct dquot *dquot)
 		mlog_errno(status);
 		goto out;
 	}
-	status = dquot_commit(dquot);
+	mutex_lock(&sb_dqopt(dquot->dq_sb)->dqio_mutex);
+	status = ocfs2_local_write_dquot(dquot);
+	mutex_unlock(&sb_dqopt(dquot->dq_sb)->dqio_mutex);
 	ocfs2_commit_trans(osb, handle);
 out:
 	mlog_exit(status);
@@ -713,6 +690,10 @@ static int ocfs2_release_dquot(struct dquot *dquot)
 
 	mlog_entry("id=%u, type=%d", dquot->dq_id, dquot->dq_type);
 
+	mutex_lock(&dquot->dq_lock);
+	/* Check whether we are not racing with some other dqget() */
+	if (atomic_read(&dquot->dq_count) > 1)
+		goto out;
 	status = ocfs2_lock_global_qf(oinfo, 1);
 	if (status < 0)
 		goto out;
@@ -723,30 +704,113 @@ static int ocfs2_release_dquot(struct dquot *dquot)
 		mlog_errno(status);
 		goto out_ilock;
 	}
-	status = dquot_release(dquot);
+
+	status = ocfs2_global_release_dquot(dquot);
+	if (status < 0) {
+		mlog_errno(status);
+		goto out_trans;
+	}
+	status = ocfs2_local_release_dquot(handle, dquot);
+	/*
+	 * If we fail here, we cannot do much as global structure is
+	 * already released. So just complain...
+	 */
+	if (status < 0)
+		mlog_errno(status);
+	clear_bit(DQ_ACTIVE_B, &dquot->dq_flags);
+out_trans:
 	ocfs2_commit_trans(osb, handle);
 out_ilock:
 	ocfs2_unlock_global_qf(oinfo, 1);
 out:
+	mutex_unlock(&dquot->dq_lock);
 	mlog_exit(status);
 	return status;
 }
 
+/*
+ * Read global dquot structure from disk or create it if it does
+ * not exist. Also update use count of the global structure and
+ * create structure in node-local quota file.
+ */
 static int ocfs2_acquire_dquot(struct dquot *dquot)
 {
-	struct ocfs2_mem_dqinfo *oinfo =
-			sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
-	int status = 0;
+	int status = 0, err;
+	int ex = 0;
+	struct super_block *sb = dquot->dq_sb;
+	struct ocfs2_super *osb = OCFS2_SB(sb);
+	int type = dquot->dq_type;
+	struct ocfs2_mem_dqinfo *info = sb_dqinfo(sb, type)->dqi_priv;
+	struct inode *gqinode = info->dqi_gqinode;
+	int need_alloc = ocfs2_global_qinit_alloc(sb, type);
+	handle_t *handle;
 
-	mlog_entry("id=%u, type=%d", dquot->dq_id, dquot->dq_type);
-	/* We need an exclusive lock, because we're going to update use count
-	 * and instantiate possibly new dquot structure */
-	status = ocfs2_lock_global_qf(oinfo, 1);
+	mlog_entry("id=%u, type=%d", dquot->dq_id, type);
+	mutex_lock(&dquot->dq_lock);
+	/*
+	 * We need an exclusive lock, because we're going to update use count
+	 * and instantiate possibly new dquot structure
+	 */
+	status = ocfs2_lock_global_qf(info, 1);
 	if (status < 0)
 		goto out;
-	status = dquot_acquire(dquot);
-	ocfs2_unlock_global_qf(oinfo, 1);
+	if (!test_bit(DQ_READ_B, &dquot->dq_flags)) {
+		status = ocfs2_qinfo_lock(info, 0);
+		if (status < 0)
+			goto out_dq;
+		status = qtree_read_dquot(&info->dqi_gi, dquot);
+		ocfs2_qinfo_unlock(info, 0);
+		if (status < 0)
+			goto out_dq;
+	}
+	set_bit(DQ_READ_B, &dquot->dq_flags);
+
+	OCFS2_DQUOT(dquot)->dq_use_count++;
+	OCFS2_DQUOT(dquot)->dq_origspace = dquot->dq_dqb.dqb_curspace;
+	OCFS2_DQUOT(dquot)->dq_originodes = dquot->dq_dqb.dqb_curinodes;
+	if (!dquot->dq_off) {	/* No real quota entry? */
+		ex = 1;
+		/*
+		 * Add blocks to quota file before we start a transaction since
+		 * locking allocators ranks above a transaction start
+		 */
+		WARN_ON(journal_current_handle());
+		status = ocfs2_extend_no_holes(gqinode,
+			gqinode->i_size + (need_alloc << sb->s_blocksize_bits),
+			gqinode->i_size);
+		if (status < 0)
+			goto out_dq;
+	}
+
+	handle = ocfs2_start_trans(osb,
+				   ocfs2_calc_global_qinit_credits(sb, type));
+	if (IS_ERR(handle)) {
+		status = PTR_ERR(handle);
+		goto out_dq;
+	}
+	status = ocfs2_qinfo_lock(info, ex);
+	if (status < 0)
+		goto out_trans;
+	status = qtree_write_dquot(&info->dqi_gi, dquot);
+	if (ex && info_dirty(sb_dqinfo(sb, type))) {
+		err = __ocfs2_global_write_info(sb, type);
+		if (!status)
+			status = err;
+	}
+	ocfs2_qinfo_unlock(info, ex);
+out_trans:
+	ocfs2_commit_trans(osb, handle);
+out_dq:
+	ocfs2_unlock_global_qf(info, 1);
+	if (status < 0)
+		goto out;
+
+	status = ocfs2_create_local_dquot(dquot);
+	if (status < 0)
+		goto out;
+	set_bit(DQ_ACTIVE_B, &dquot->dq_flags);
 out:
+	mutex_unlock(&dquot->dq_lock);
 	mlog_exit(status);
 	return status;
 }
@@ -768,7 +832,6 @@ static int ocfs2_mark_dquot_dirty(struct dquot *dquot)
 	struct ocfs2_super *osb = OCFS2_SB(sb);
 
 	mlog_entry("id=%u, type=%d", dquot->dq_id, type);
-	dquot_mark_dquot_dirty(dquot);
 
 	/* In case user set some limits, sync dquot immediately to global
 	 * quota file so that information propagates quicker */
@@ -791,14 +854,16 @@ static int ocfs2_mark_dquot_dirty(struct dquot *dquot)
 		mlog_errno(status);
 		goto out_ilock;
 	}
+	mutex_lock(&sb_dqopt(sb)->dqio_mutex);
 	status = ocfs2_sync_dquot(dquot);
 	if (status < 0) {
 		mlog_errno(status);
-		goto out_trans;
+		goto out_dlock;
 	}
 	/* Now write updated local dquot structure */
-	status = dquot_commit(dquot);
-out_trans:
+	status = ocfs2_local_write_dquot(dquot);
+out_dlock:
+	mutex_unlock(&sb_dqopt(sb)->dqio_mutex);
 	ocfs2_commit_trans(osb, handle);
 out_ilock:
 	ocfs2_unlock_global_qf(oinfo, 1);
@@ -850,7 +915,7 @@ static void ocfs2_destroy_dquot(struct dquot *dquot)
 }
 
 const struct dquot_operations ocfs2_quota_operations = {
-	.write_dquot	= ocfs2_write_dquot,
+	/* We never make dquot dirty so .write_dquot is never called */
 	.acquire_dquot	= ocfs2_acquire_dquot,
 	.release_dquot	= ocfs2_release_dquot,
 	.mark_dirty	= ocfs2_mark_dquot_dirty,

fs/ocfs2/quota_local.c

@@ -22,6 +22,7 @@
 #include "dlmglue.h"
 #include "quota.h"
 #include "uptodate.h"
+#include "super.h"
 
 /* Number of local quota structures per block */
 static inline unsigned int ol_quota_entries_per_block(struct super_block *sb)
@@ -129,6 +130,39 @@ static int ocfs2_modify_bh(struct inode *inode, struct buffer_head *bh,
 	return 0;
 }
 
+/*
+ * Read quota block from a given logical offset.
+ *
+ * This function acquires ip_alloc_sem and thus it must not be called with a
+ * transaction started.
+ */
+static int ocfs2_read_quota_block(struct inode *inode, u64 v_block,
+				  struct buffer_head **bh)
+{
+	int rc = 0;
+	struct buffer_head *tmp = *bh;
+
+	if (i_size_read(inode) >> inode->i_sb->s_blocksize_bits <= v_block) {
+		ocfs2_error(inode->i_sb,
+			    "Quota file %llu is probably corrupted! Requested "
+			    "to read block %Lu but file has size only %Lu\n",
+			    (unsigned long long)OCFS2_I(inode)->ip_blkno,
+			    (unsigned long long)v_block,
+			    (unsigned long long)i_size_read(inode));
+		return -EIO;
+	}
+	rc = ocfs2_read_virt_blocks(inode, v_block, 1, &tmp, 0,
+				    ocfs2_validate_quota_block);
+	if (rc)
+		mlog_errno(rc);
+
+	/* If ocfs2_read_virt_blocks() got us a new bh, pass it up. */
+	if (!rc && !*bh)
+		*bh = tmp;
+
+	return rc;
+}
+
 /* Check whether we understand format of quota files */
 static int ocfs2_local_check_quota_file(struct super_block *sb, int type)
 {
@@ -671,7 +705,7 @@ static int ocfs2_local_read_info(struct super_block *sb, int type)
 	INIT_LIST_HEAD(&oinfo->dqi_chunk);
 	oinfo->dqi_rec = NULL;
 	oinfo->dqi_lqi_bh = NULL;
-	oinfo->dqi_ibh = NULL;
+	oinfo->dqi_libh = NULL;
 
 	status = ocfs2_global_read_info(sb, type);
 	if (status < 0)
@@ -697,7 +731,7 @@ static int ocfs2_local_read_info(struct super_block *sb, int type)
 	info->dqi_flags = le32_to_cpu(ldinfo->dqi_flags);
 	oinfo->dqi_chunks = le32_to_cpu(ldinfo->dqi_chunks);
 	oinfo->dqi_blocks = le32_to_cpu(ldinfo->dqi_blocks);
-	oinfo->dqi_ibh = bh;
+	oinfo->dqi_libh = bh;
 
 	/* We crashed when using local quota file? */
 	if (!(info->dqi_flags & OLQF_CLEAN)) {
@@ -759,7 +793,7 @@ static int ocfs2_local_write_info(struct super_block *sb, int type)
 {
 	struct mem_dqinfo *info = sb_dqinfo(sb, type);
 	struct buffer_head *bh = ((struct ocfs2_mem_dqinfo *)info->dqi_priv)
-						->dqi_ibh;
+						->dqi_libh;
 	int status;
 
 	status = ocfs2_modify_bh(sb_dqopt(sb)->files[type], bh, olq_update_info,
@@ -782,10 +816,6 @@ static int ocfs2_local_free_info(struct super_block *sb, int type)
 	int mark_clean = 1, len;
 	int status;
 
-	/* At this point we know there are no more dquots and thus
-	 * even if there's some sync in the pdflush queue, it won't
-	 * find any dquots and return without doing anything */
-	cancel_delayed_work_sync(&oinfo->dqi_sync_work);
 	iput(oinfo->dqi_gqinode);
 	ocfs2_simple_drop_lockres(OCFS2_SB(sb), &oinfo->dqi_gqlock);
 	ocfs2_lock_res_free(&oinfo->dqi_gqlock);
@@ -820,7 +850,7 @@ static int ocfs2_local_free_info(struct super_block *sb, int type)
 	/* Mark local file as clean */
 	info->dqi_flags |= OLQF_CLEAN;
 	status = ocfs2_modify_bh(sb_dqopt(sb)->files[type],
-				 oinfo->dqi_ibh,
+				 oinfo->dqi_libh,
 				 olq_update_info,
 				 info);
 	if (status < 0) {
@@ -830,7 +860,7 @@ static int ocfs2_local_free_info(struct super_block *sb, int type)
 
 out:
 	ocfs2_inode_unlock(sb_dqopt(sb)->files[type], 1);
-	brelse(oinfo->dqi_ibh);
+	brelse(oinfo->dqi_libh);
 	brelse(oinfo->dqi_lqi_bh);
 	kfree(oinfo);
 	return 0;
@@ -858,22 +888,21 @@ static void olq_set_dquot(struct buffer_head *bh, void *private)
 }
 
 /* Write dquot to local quota file */
-static int ocfs2_local_write_dquot(struct dquot *dquot)
+int ocfs2_local_write_dquot(struct dquot *dquot)
 {
 	struct super_block *sb = dquot->dq_sb;
 	struct ocfs2_dquot *od = OCFS2_DQUOT(dquot);
-	struct buffer_head *bh = NULL;
+	struct buffer_head *bh;
+	struct inode *lqinode = sb_dqopt(sb)->files[dquot->dq_type];
 	int status;
 
-	status = ocfs2_read_quota_block(sb_dqopt(sb)->files[dquot->dq_type],
-				    ol_dqblk_file_block(sb, od->dq_local_off),
-				    &bh);
+	status = ocfs2_read_quota_phys_block(lqinode, od->dq_local_phys_blk,
+					     &bh);
 	if (status) {
 		mlog_errno(status);
 		goto out;
 	}
-	status = ocfs2_modify_bh(sb_dqopt(sb)->files[dquot->dq_type], bh,
-				 olq_set_dquot, od);
+	status = ocfs2_modify_bh(lqinode, bh, olq_set_dquot, od);
 	if (status < 0) {
 		mlog_errno(status);
 		goto out;
@@ -973,10 +1002,8 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk(
 	}
 
 	/* Initialize chunk header */
-	down_read(&OCFS2_I(lqinode)->ip_alloc_sem);
 	status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks,
 					     &p_blkno, NULL, NULL);
-	up_read(&OCFS2_I(lqinode)->ip_alloc_sem);
 	if (status < 0) {
 		mlog_errno(status);
 		goto out_trans;
@@ -1004,10 +1031,8 @@ static struct ocfs2_quota_chunk *ocfs2_local_quota_add_chunk(
 	ocfs2_journal_dirty(handle, bh);
 
 	/* Initialize new block with structures */
-	down_read(&OCFS2_I(lqinode)->ip_alloc_sem);
 	status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks + 1,
 					     &p_blkno, NULL, NULL);
-	up_read(&OCFS2_I(lqinode)->ip_alloc_sem);
 	if (status < 0) {
 		mlog_errno(status);
 		goto out_trans;
@@ -1104,10 +1129,8 @@ static struct ocfs2_quota_chunk *ocfs2_extend_local_quota_file(
 	}
 
 	/* Get buffer from the just added block */
-	down_read(&OCFS2_I(lqinode)->ip_alloc_sem);
 	status = ocfs2_extent_map_get_blocks(lqinode, oinfo->dqi_blocks,
 					     &p_blkno, NULL, NULL);
-	up_read(&OCFS2_I(lqinode)->ip_alloc_sem);
 	if (status < 0) {
 		mlog_errno(status);
 		goto out;
@@ -1188,7 +1211,7 @@ static void olq_alloc_dquot(struct buffer_head *bh, void *private)
 }
 
 /* Create dquot in the local file for given id */
-static int ocfs2_create_local_dquot(struct dquot *dquot)
+int ocfs2_create_local_dquot(struct dquot *dquot)
 {
 	struct super_block *sb = dquot->dq_sb;
 	int type = dquot->dq_type;
@@ -1197,17 +1220,27 @@ static int ocfs2_create_local_dquot(struct dquot *dquot)
 	struct ocfs2_dquot *od = OCFS2_DQUOT(dquot);
 	int offset;
 	int status;
+	u64 pcount;
 
+	down_write(&OCFS2_I(lqinode)->ip_alloc_sem);
 	chunk = ocfs2_find_free_entry(sb, type, &offset);
 	if (!chunk) {
 		chunk = ocfs2_extend_local_quota_file(sb, type, &offset);
-		if (IS_ERR(chunk))
-			return PTR_ERR(chunk);
+		if (IS_ERR(chunk)) {
+			status = PTR_ERR(chunk);
+			goto out;
+		}
 	} else if (IS_ERR(chunk)) {
-		return PTR_ERR(chunk);
+		status = PTR_ERR(chunk);
+		goto out;
 	}
 	od->dq_local_off = ol_dqblk_off(sb, chunk->qc_num, offset);
 	od->dq_chunk = chunk;
+	status = ocfs2_extent_map_get_blocks(lqinode,
+				     ol_dqblk_block(sb, chunk->qc_num, offset),
+				     &od->dq_local_phys_blk,
+				     &pcount,
+				     NULL);
 
 	/* Initialize dquot structure on disk */
 	status = ocfs2_local_write_dquot(dquot);
@@ -1224,39 +1257,15 @@ static int ocfs2_create_local_dquot(struct dquot *dquot)
 		goto out;
 	}
 out:
+	up_write(&OCFS2_I(lqinode)->ip_alloc_sem);
 	return status;
 }
 
-/* Create entry in local file for dquot, load data from the global file */
-static int ocfs2_local_read_dquot(struct dquot *dquot)
-{
-	int status;
-
-	mlog_entry("id=%u, type=%d\n", dquot->dq_id, dquot->dq_type);
-
-	status = ocfs2_global_read_dquot(dquot);
-	if (status < 0) {
-		mlog_errno(status);
-		goto out_err;
-	}
-
-	/* Now create entry in the local quota file */
-	status = ocfs2_create_local_dquot(dquot);
-	if (status < 0) {
-		mlog_errno(status);
-		goto out_err;
-	}
-	mlog_exit(0);
-	return 0;
-out_err:
-	mlog_exit(status);
-	return status;
-}
-
-/* Release dquot structure from local quota file. ocfs2_release_dquot() has
- * already started a transaction and obtained exclusive lock for global
- * quota file. */
-static int ocfs2_local_release_dquot(struct dquot *dquot)
+/*
+ * Release dquot structure from local quota file. ocfs2_release_dquot() has
+ * already started a transaction and written all changes to global quota file
+ */
+int ocfs2_local_release_dquot(handle_t *handle, struct dquot *dquot)
 {
 	int status;
 	int type = dquot->dq_type;
@@ -1264,15 +1273,6 @@ static int ocfs2_local_release_dquot(struct dquot *dquot)
 	struct super_block *sb = dquot->dq_sb;
 	struct ocfs2_local_disk_chunk *dchunk;
 	int offset;
-	handle_t *handle = journal_current_handle();
-
-	BUG_ON(!handle);
-	/* First write all local changes to global file */
-	status = ocfs2_global_release_dquot(dquot);
-	if (status < 0) {
-		mlog_errno(status);
-		goto out;
-	}
 
 	status = ocfs2_journal_access_dq(handle,
 			INODE_CACHE(sb_dqopt(sb)->files[type]),
@@ -1305,9 +1305,6 @@ static const struct quota_format_ops ocfs2_format_ops = {
 	.read_file_info		= ocfs2_local_read_info,
 	.write_file_info	= ocfs2_global_write_info,
 	.free_file_info		= ocfs2_local_free_info,
-	.read_dqblk		= ocfs2_local_read_dquot,
-	.commit_dqblk		= ocfs2_local_write_dquot,
-	.release_dqblk		= ocfs2_local_release_dquot,
 };
 
 struct quota_format_type ocfs2_quota_format = {

fs/ocfs2/super.c

@@ -938,12 +938,16 @@ static void ocfs2_disable_quotas(struct ocfs2_super *osb)
 	int type;
 	struct inode *inode;
 	struct super_block *sb = osb->sb;
+	struct ocfs2_mem_dqinfo *oinfo;
 
 	/* We mostly ignore errors in this function because there's not much
 	 * we can do when we see them */
 	for (type = 0; type < MAXQUOTAS; type++) {
 		if (!sb_has_quota_loaded(sb, type))
 			continue;
+		/* Cancel periodic syncing before we grab dqonoff_mutex */
+		oinfo = sb_dqinfo(sb, type)->dqi_priv;
+		cancel_delayed_work_sync(&oinfo->dqi_sync_work);
 		inode = igrab(sb->s_dquot.files[type]);
 		/* Turn off quotas. This will remove all dquot structures from
 		 * memory and so they will be automatically synced to global

fs/quota/dquot.c

@@ -82,7 +82,7 @@
 
 /*
  * There are three quota SMP locks. dq_list_lock protects all lists with quotas
- * and quota formats, dqstats structure containing statistics about the lists
+ * and quota formats.
  * dq_data_lock protects data from dq_dqb and also mem_dqinfo structures and
  * also guards consistency of dquot->dq_dqb with inode->i_blocks, i_bytes.
  * i_blocks and i_bytes updates itself are guarded by i_lock acquired directly
@@ -132,7 +132,9 @@ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_state_lock);
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_data_lock);
 EXPORT_SYMBOL(dq_data_lock);
 
+#if defined(CONFIG_QUOTA_DEBUG) || defined(CONFIG_PRINT_QUOTA_WARNING)
 static char *quotatypes[] = INITQFNAMES;
+#endif
 static struct quota_format_type *quota_formats;	/* List of registered formats */
 static struct quota_module_name module_names[] = INIT_QUOTA_MODULE_NAMES;
 
@@ -226,6 +228,10 @@ static struct hlist_head *dquot_hash;
 
 struct dqstats dqstats;
 EXPORT_SYMBOL(dqstats);
+#ifdef CONFIG_SMP
+struct dqstats *dqstats_pcpu;
+EXPORT_SYMBOL(dqstats_pcpu);
+#endif
 
 static qsize_t inode_get_rsv_space(struct inode *inode);
 static void __dquot_initialize(struct inode *inode, int type);
@@ -273,7 +279,7 @@ static struct dquot *find_dquot(unsigned int hashent, struct super_block *sb,
 static inline void put_dquot_last(struct dquot *dquot)
 {
 	list_add_tail(&dquot->dq_free, &free_dquots);
-	dqstats.free_dquots++;
+	dqstats_inc(DQST_FREE_DQUOTS);
 }
 
 static inline void remove_free_dquot(struct dquot *dquot)
@@ -281,7 +287,7 @@ static inline void remove_free_dquot(struct dquot *dquot)
 	if (list_empty(&dquot->dq_free))
 		return;
 	list_del_init(&dquot->dq_free);
-	dqstats.free_dquots--;
+	dqstats_dec(DQST_FREE_DQUOTS);
 }
 
 static inline void put_inuse(struct dquot *dquot)
@@ -289,12 +295,12 @@ static inline void put_inuse(struct dquot *dquot)
 	/* We add to the back of inuse list so we don't have to restart
 	 * when traversing this list and we block */
 	list_add_tail(&dquot->dq_inuse, &inuse_list);
-	dqstats.allocated_dquots++;
+	dqstats_inc(DQST_ALLOC_DQUOTS);
 }
 
 static inline void remove_inuse(struct dquot *dquot)
 {
-	dqstats.allocated_dquots--;
+	dqstats_dec(DQST_ALLOC_DQUOTS);
 	list_del(&dquot->dq_inuse);
 }
 /*
@@ -317,14 +323,23 @@ static inline int mark_dquot_dirty(struct dquot *dquot)
 	return dquot->dq_sb->dq_op->mark_dirty(dquot);
 }
 
+/* Mark dquot dirty in atomic manner, and return it's old dirty flag state */
 int dquot_mark_dquot_dirty(struct dquot *dquot)
 {
+	int ret = 1;
+
+	/* If quota is dirty already, we don't have to acquire dq_list_lock */
+	if (test_bit(DQ_MOD_B, &dquot->dq_flags))
+		return 1;
+
 	spin_lock(&dq_list_lock);
-	if (!test_and_set_bit(DQ_MOD_B, &dquot->dq_flags))
+	if (!test_and_set_bit(DQ_MOD_B, &dquot->dq_flags)) {
 		list_add(&dquot->dq_dirty, &sb_dqopt(dquot->dq_sb)->
 				info[dquot->dq_type].dqi_dirty_list);
+		ret = 0;
+	}
 	spin_unlock(&dq_list_lock);
-	return 0;
+	return ret;
 }
 EXPORT_SYMBOL(dquot_mark_dquot_dirty);
 
@@ -550,8 +565,8 @@ int dquot_scan_active(struct super_block *sb,
 			continue;
 		/* Now we have active dquot so we can just increase use count */
 		atomic_inc(&dquot->dq_count);
-		dqstats.lookups++;
 		spin_unlock(&dq_list_lock);
+		dqstats_inc(DQST_LOOKUPS);
 		dqput(old_dquot);
 		old_dquot = dquot;
 		ret = fn(dquot, priv);
@@ -596,8 +611,8 @@ int vfs_quota_sync(struct super_block *sb, int type, int wait)
  			 * holding reference so we can safely just increase
 			 * use count */
 			atomic_inc(&dquot->dq_count);
-			dqstats.lookups++;
 			spin_unlock(&dq_list_lock);
+			dqstats_inc(DQST_LOOKUPS);
 			sb->dq_op->write_dquot(dquot);
 			dqput(dquot);
 			spin_lock(&dq_list_lock);
@@ -609,9 +624,7 @@ int vfs_quota_sync(struct super_block *sb, int type, int wait)
 		if ((cnt == type || type == -1) && sb_has_quota_active(sb, cnt)
 		    && info_dirty(&dqopt->info[cnt]))
 			sb->dq_op->write_info(sb, cnt);
-	spin_lock(&dq_list_lock);
-	dqstats.syncs++;
-	spin_unlock(&dq_list_lock);
+	dqstats_inc(DQST_SYNCS);
 	mutex_unlock(&dqopt->dqonoff_mutex);
 
 	if (!wait || (sb_dqopt(sb)->flags & DQUOT_QUOTA_SYS_FILE))
@@ -663,6 +676,22 @@ static void prune_dqcache(int count)
 	}
 }
 
+static int dqstats_read(unsigned int type)
+{
+	int count = 0;
+#ifdef CONFIG_SMP
+	int cpu;
+	for_each_possible_cpu(cpu)
+		count += per_cpu_ptr(dqstats_pcpu, cpu)->stat[type];
+	/* Statistics reading is racy, but absolute accuracy isn't required */
+	if (count < 0)
+		count = 0;
+#else
+	count = dqstats.stat[type];
+#endif
+	return count;
+}
+
 /*
  * This is called from kswapd when we think we need some
  * more memory
@@ -675,7 +704,7 @@ static int shrink_dqcache_memory(int nr, gfp_t gfp_mask)
 		prune_dqcache(nr);
 		spin_unlock(&dq_list_lock);
 	}
-	return (dqstats.free_dquots / 100) * sysctl_vfs_cache_pressure;
+	return (dqstats_read(DQST_FREE_DQUOTS)/100) * sysctl_vfs_cache_pressure;
 }
 
 static struct shrinker dqcache_shrinker = {
@@ -703,10 +732,7 @@ void dqput(struct dquot *dquot)
 		BUG();
 	}
 #endif
-	
-	spin_lock(&dq_list_lock);
-	dqstats.drops++;
-	spin_unlock(&dq_list_lock);
+	dqstats_inc(DQST_DROPS);
 we_slept:
 	spin_lock(&dq_list_lock);
 	if (atomic_read(&dquot->dq_count) > 1) {
@@ -823,15 +849,15 @@ we_slept:
 		put_inuse(dquot);
 		/* hash it first so it can be found */
 		insert_dquot_hash(dquot);
-		dqstats.lookups++;
 		spin_unlock(&dq_list_lock);
+		dqstats_inc(DQST_LOOKUPS);
 	} else {
 		if (!atomic_read(&dquot->dq_count))
 			remove_free_dquot(dquot);
 		atomic_inc(&dquot->dq_count);
-		dqstats.cache_hits++;
-		dqstats.lookups++;
 		spin_unlock(&dq_list_lock);
+		dqstats_inc(DQST_CACHE_HITS);
+		dqstats_inc(DQST_LOOKUPS);
 	}
 	/* Wait for dq_lock - after this we know that either dquot_release() is
 	 * already finished or it will be canceled due to dq_count > 1 test */
@@ -1677,16 +1703,19 @@ EXPORT_SYMBOL(dquot_free_inode);
 
 /*
  * Transfer the number of inode and blocks from one diskquota to an other.
+ * On success, dquot references in transfer_to are consumed and references
+ * to original dquots that need to be released are placed there. On failure,
+ * references are kept untouched.
  *
  * This operation can block, but only after everything is updated
  * A transaction must be started when entering this function.
+ *
  */
-static int __dquot_transfer(struct inode *inode, qid_t *chid, unsigned long mask)
+int __dquot_transfer(struct inode *inode, struct dquot **transfer_to)
 {
 	qsize_t space, cur_space;
 	qsize_t rsv_space = 0;
-	struct dquot *transfer_from[MAXQUOTAS];
-	struct dquot *transfer_to[MAXQUOTAS];
+	struct dquot *transfer_from[MAXQUOTAS] = {};
 	int cnt, ret = 0;
 	char warntype_to[MAXQUOTAS];
 	char warntype_from_inodes[MAXQUOTAS], warntype_from_space[MAXQUOTAS];
@@ -1696,19 +1725,12 @@ static int __dquot_transfer(struct inode *inode, qid_t *chid, unsigned long mask
 	if (IS_NOQUOTA(inode))
 		return 0;
 	/* Initialize the arrays */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		transfer_from[cnt] = NULL;
-		transfer_to[cnt] = NULL;
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
 		warntype_to[cnt] = QUOTA_NL_NOWARN;
-	}
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		if (mask & (1 << cnt))
-			transfer_to[cnt] = dqget(inode->i_sb, chid[cnt], cnt);
-	}
 	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	if (IS_NOQUOTA(inode)) {	/* File without quota accounting? */
 		up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
-		goto put_all;
+		return 0;
 	}
 	spin_lock(&dq_data_lock);
 	cur_space = inode_get_bytes(inode);
@@ -1760,47 +1782,41 @@ static int __dquot_transfer(struct inode *inode, qid_t *chid, unsigned long mask
 
 	mark_all_dquot_dirty(transfer_from);
 	mark_all_dquot_dirty(transfer_to);
-	/* The reference we got is transferred to the inode */
+	/* Pass back references to put */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		transfer_to[cnt] = NULL;
-warn_put_all:
+		transfer_to[cnt] = transfer_from[cnt];
+warn:
 	flush_warnings(transfer_to, warntype_to);
 	flush_warnings(transfer_from, warntype_from_inodes);
 	flush_warnings(transfer_from, warntype_from_space);
-put_all:
-	dqput_all(transfer_from);
-	dqput_all(transfer_to);
 	return ret;
 over_quota:
 	spin_unlock(&dq_data_lock);
 	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
-	/* Clear dquot pointers we don't want to dqput() */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		transfer_from[cnt] = NULL;
-	goto warn_put_all;
+	goto warn;
 }
+EXPORT_SYMBOL(__dquot_transfer);
 
 /* Wrapper for transferring ownership of an inode for uid/gid only
  * Called from FSXXX_setattr()
  */
 int dquot_transfer(struct inode *inode, struct iattr *iattr)
 {
-	qid_t chid[MAXQUOTAS];
-	unsigned long mask = 0;
+	struct dquot *transfer_to[MAXQUOTAS] = {};
+	struct super_block *sb = inode->i_sb;
+	int ret;
 
-	if (iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid) {
-		mask |= 1 << USRQUOTA;
-		chid[USRQUOTA] = iattr->ia_uid;
-	}
-	if (iattr->ia_valid & ATTR_GID && iattr->ia_gid != inode->i_gid) {
-		mask |= 1 << GRPQUOTA;
-		chid[GRPQUOTA] = iattr->ia_gid;
-	}
-	if (sb_any_quota_active(inode->i_sb) && !IS_NOQUOTA(inode)) {
-		dquot_initialize(inode);
-		return __dquot_transfer(inode, chid, mask);
-	}
-	return 0;
+	if (!sb_any_quota_active(sb) || IS_NOQUOTA(inode))
+		return 0;
+
+	if (iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid)
+		transfer_to[USRQUOTA] = dqget(sb, iattr->ia_uid, USRQUOTA);
+	if (iattr->ia_valid & ATTR_GID && iattr->ia_gid != inode->i_gid)
+		transfer_to[GRPQUOTA] = dqget(sb, iattr->ia_uid, GRPQUOTA);
+
+	ret = __dquot_transfer(inode, transfer_to);
+	dqput_all(transfer_to);
+	return ret;
 }
 EXPORT_SYMBOL(dquot_transfer);
 
@@ -2275,25 +2291,30 @@ static inline qsize_t stoqb(qsize_t space)
 }
 
 /* Generic routine for getting common part of quota structure */
-static void do_get_dqblk(struct dquot *dquot, struct if_dqblk *di)
+static void do_get_dqblk(struct dquot *dquot, struct fs_disk_quota *di)
 {
 	struct mem_dqblk *dm = &dquot->dq_dqb;
 
+	memset(di, 0, sizeof(*di));
+	di->d_version = FS_DQUOT_VERSION;
+	di->d_flags = dquot->dq_type == USRQUOTA ?
+			XFS_USER_QUOTA : XFS_GROUP_QUOTA;
+	di->d_id = dquot->dq_id;
+
 	spin_lock(&dq_data_lock);
-	di->dqb_bhardlimit = stoqb(dm->dqb_bhardlimit);
-	di->dqb_bsoftlimit = stoqb(dm->dqb_bsoftlimit);
-	di->dqb_curspace = dm->dqb_curspace + dm->dqb_rsvspace;
-	di->dqb_ihardlimit = dm->dqb_ihardlimit;
-	di->dqb_isoftlimit = dm->dqb_isoftlimit;
-	di->dqb_curinodes = dm->dqb_curinodes;
-	di->dqb_btime = dm->dqb_btime;
-	di->dqb_itime = dm->dqb_itime;
-	di->dqb_valid = QIF_ALL;
+	di->d_blk_hardlimit = stoqb(dm->dqb_bhardlimit);
+	di->d_blk_softlimit = stoqb(dm->dqb_bsoftlimit);
+	di->d_ino_hardlimit = dm->dqb_ihardlimit;
+	di->d_ino_softlimit = dm->dqb_isoftlimit;
+	di->d_bcount = dm->dqb_curspace + dm->dqb_rsvspace;
+	di->d_icount = dm->dqb_curinodes;
+	di->d_btimer = dm->dqb_btime;
+	di->d_itimer = dm->dqb_itime;
 	spin_unlock(&dq_data_lock);
 }
 
 int vfs_get_dqblk(struct super_block *sb, int type, qid_t id,
-		  struct if_dqblk *di)
+		  struct fs_disk_quota *di)
 {
 	struct dquot *dquot;
 
@@ -2307,51 +2328,70 @@ int vfs_get_dqblk(struct super_block *sb, int type, qid_t id,
 }
 EXPORT_SYMBOL(vfs_get_dqblk);
 
+#define VFS_FS_DQ_MASK \
+	(FS_DQ_BCOUNT | FS_DQ_BSOFT | FS_DQ_BHARD | \
+	 FS_DQ_ICOUNT | FS_DQ_ISOFT | FS_DQ_IHARD | \
+	 FS_DQ_BTIMER | FS_DQ_ITIMER)
+
 /* Generic routine for setting common part of quota structure */
-static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di)
+static int do_set_dqblk(struct dquot *dquot, struct fs_disk_quota *di)
 {
 	struct mem_dqblk *dm = &dquot->dq_dqb;
 	int check_blim = 0, check_ilim = 0;
 	struct mem_dqinfo *dqi = &sb_dqopt(dquot->dq_sb)->info[dquot->dq_type];
 
-	if ((di->dqb_valid & QIF_BLIMITS &&
-	     (di->dqb_bhardlimit > dqi->dqi_maxblimit ||
-	      di->dqb_bsoftlimit > dqi->dqi_maxblimit)) ||
-	    (di->dqb_valid & QIF_ILIMITS &&
-	     (di->dqb_ihardlimit > dqi->dqi_maxilimit ||
-	      di->dqb_isoftlimit > dqi->dqi_maxilimit)))
+	if (di->d_fieldmask & ~VFS_FS_DQ_MASK)
+		return -EINVAL;
+
+	if (((di->d_fieldmask & FS_DQ_BSOFT) &&
+	     (di->d_blk_softlimit > dqi->dqi_maxblimit)) ||
+	    ((di->d_fieldmask & FS_DQ_BHARD) &&
+	     (di->d_blk_hardlimit > dqi->dqi_maxblimit)) ||
+	    ((di->d_fieldmask & FS_DQ_ISOFT) &&
+	     (di->d_ino_softlimit > dqi->dqi_maxilimit)) ||
+	    ((di->d_fieldmask & FS_DQ_IHARD) &&
+	     (di->d_ino_hardlimit > dqi->dqi_maxilimit)))
 		return -ERANGE;
 
 	spin_lock(&dq_data_lock);
-	if (di->dqb_valid & QIF_SPACE) {
-		dm->dqb_curspace = di->dqb_curspace - dm->dqb_rsvspace;
+	if (di->d_fieldmask & FS_DQ_BCOUNT) {
+		dm->dqb_curspace = di->d_bcount - dm->dqb_rsvspace;
 		check_blim = 1;
 		set_bit(DQ_LASTSET_B + QIF_SPACE_B, &dquot->dq_flags);
 	}
-	if (di->dqb_valid & QIF_BLIMITS) {
-		dm->dqb_bsoftlimit = qbtos(di->dqb_bsoftlimit);
-		dm->dqb_bhardlimit = qbtos(di->dqb_bhardlimit);
+
+	if (di->d_fieldmask & FS_DQ_BSOFT)
+		dm->dqb_bsoftlimit = qbtos(di->d_blk_softlimit);
+	if (di->d_fieldmask & FS_DQ_BHARD)
+		dm->dqb_bhardlimit = qbtos(di->d_blk_hardlimit);
+	if (di->d_fieldmask & (FS_DQ_BSOFT | FS_DQ_BHARD)) {
 		check_blim = 1;
 		set_bit(DQ_LASTSET_B + QIF_BLIMITS_B, &dquot->dq_flags);
 	}
-	if (di->dqb_valid & QIF_INODES) {
-		dm->dqb_curinodes = di->dqb_curinodes;
+
+	if (di->d_fieldmask & FS_DQ_ICOUNT) {
+		dm->dqb_curinodes = di->d_icount;
 		check_ilim = 1;
 		set_bit(DQ_LASTSET_B + QIF_INODES_B, &dquot->dq_flags);
 	}
-	if (di->dqb_valid & QIF_ILIMITS) {
-		dm->dqb_isoftlimit = di->dqb_isoftlimit;
-		dm->dqb_ihardlimit = di->dqb_ihardlimit;
+
+	if (di->d_fieldmask & FS_DQ_ISOFT)
+		dm->dqb_isoftlimit = di->d_ino_softlimit;
+	if (di->d_fieldmask & FS_DQ_IHARD)
+		dm->dqb_ihardlimit = di->d_ino_hardlimit;
+	if (di->d_fieldmask & (FS_DQ_ISOFT | FS_DQ_IHARD)) {
 		check_ilim = 1;
 		set_bit(DQ_LASTSET_B + QIF_ILIMITS_B, &dquot->dq_flags);
 	}
-	if (di->dqb_valid & QIF_BTIME) {
-		dm->dqb_btime = di->dqb_btime;
+
+	if (di->d_fieldmask & FS_DQ_BTIMER) {
+		dm->dqb_btime = di->d_btimer;
 		check_blim = 1;
 		set_bit(DQ_LASTSET_B + QIF_BTIME_B, &dquot->dq_flags);
 	}
-	if (di->dqb_valid & QIF_ITIME) {
-		dm->dqb_itime = di->dqb_itime;
+
+	if (di->d_fieldmask & FS_DQ_ITIMER) {
+		dm->dqb_itime = di->d_itimer;
 		check_ilim = 1;
 		set_bit(DQ_LASTSET_B + QIF_ITIME_B, &dquot->dq_flags);
 	}
@@ -2361,7 +2401,7 @@ static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di)
 		    dm->dqb_curspace < dm->dqb_bsoftlimit) {
 			dm->dqb_btime = 0;
 			clear_bit(DQ_BLKS_B, &dquot->dq_flags);
-		} else if (!(di->dqb_valid & QIF_BTIME))
+		} else if (!(di->d_fieldmask & FS_DQ_BTIMER))
 			/* Set grace only if user hasn't provided his own... */
 			dm->dqb_btime = get_seconds() + dqi->dqi_bgrace;
 	}
@@ -2370,7 +2410,7 @@ static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di)
 		    dm->dqb_curinodes < dm->dqb_isoftlimit) {
 			dm->dqb_itime = 0;
 			clear_bit(DQ_INODES_B, &dquot->dq_flags);
-		} else if (!(di->dqb_valid & QIF_ITIME))
+		} else if (!(di->d_fieldmask & FS_DQ_ITIMER))
 			/* Set grace only if user hasn't provided his own... */
 			dm->dqb_itime = get_seconds() + dqi->dqi_igrace;
 	}
@@ -2386,7 +2426,7 @@ static int do_set_dqblk(struct dquot *dquot, struct if_dqblk *di)
 }
 
 int vfs_set_dqblk(struct super_block *sb, int type, qid_t id,
-		  struct if_dqblk *di)
+		  struct fs_disk_quota *di)
 {
 	struct dquot *dquot;
 	int rc;
@@ -2465,62 +2505,74 @@ const struct quotactl_ops vfs_quotactl_ops = {
 	.set_dqblk	= vfs_set_dqblk
 };
 
+
+static int do_proc_dqstats(struct ctl_table *table, int write,
+		     void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+#ifdef CONFIG_SMP
+	/* Update global table */
+	unsigned int type = (int *)table->data - dqstats.stat;
+	dqstats.stat[type] = dqstats_read(type);
+#endif
+	return proc_dointvec(table, write, buffer, lenp, ppos);
+}
+
 static ctl_table fs_dqstats_table[] = {
 	{
 		.procname	= "lookups",
-		.data		= &dqstats.lookups,
+		.data		= &dqstats.stat[DQST_LOOKUPS],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "drops",
-		.data		= &dqstats.drops,
+		.data		= &dqstats.stat[DQST_DROPS],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "reads",
-		.data		= &dqstats.reads,
+		.data		= &dqstats.stat[DQST_READS],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "writes",
-		.data		= &dqstats.writes,
+		.data		= &dqstats.stat[DQST_WRITES],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "cache_hits",
-		.data		= &dqstats.cache_hits,
+		.data		= &dqstats.stat[DQST_CACHE_HITS],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "allocated_dquots",
-		.data		= &dqstats.allocated_dquots,
+		.data		= &dqstats.stat[DQST_ALLOC_DQUOTS],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "free_dquots",
-		.data		= &dqstats.free_dquots,
+		.data		= &dqstats.stat[DQST_FREE_DQUOTS],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "syncs",
-		.data		= &dqstats.syncs,
+		.data		= &dqstats.stat[DQST_SYNCS],
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
-		.proc_handler	= proc_dointvec,
+		.proc_handler	= do_proc_dqstats,
 	},
 #ifdef CONFIG_PRINT_QUOTA_WARNING
 	{
@@ -2572,6 +2624,13 @@ static int __init dquot_init(void)
 	if (!dquot_hash)
 		panic("Cannot create dquot hash table");
 
+#ifdef CONFIG_SMP
+	dqstats_pcpu = alloc_percpu(struct dqstats);
+	if (!dqstats_pcpu)
+		panic("Cannot create dquot stats table");
+#endif
+	memset(&dqstats, 0, sizeof(struct dqstats));
+
 	/* Find power-of-two hlist_heads which can fit into allocation */
 	nr_hash = (1UL << order) * PAGE_SIZE / sizeof(struct hlist_head);
 	dq_hash_bits = 0;

fs/quota/quota.c

@@ -113,8 +113,6 @@ static int quota_getinfo(struct super_block *sb, int type, void __user *addr)
 	struct if_dqinfo info;
 	int ret;
 
-	if (!sb_has_quota_active(sb, type))
-		return -ESRCH;
 	if (!sb->s_qcop->get_info)
 		return -ENOSYS;
 	ret = sb->s_qcop->get_info(sb, type, &info);
@@ -129,43 +127,80 @@ static int quota_setinfo(struct super_block *sb, int type, void __user *addr)
 
 	if (copy_from_user(&info, addr, sizeof(info)))
 		return -EFAULT;
-	if (!sb_has_quota_active(sb, type))
-		return -ESRCH;
 	if (!sb->s_qcop->set_info)
 		return -ENOSYS;
 	return sb->s_qcop->set_info(sb, type, &info);
 }
 
+static void copy_to_if_dqblk(struct if_dqblk *dst, struct fs_disk_quota *src)
+{
+	dst->dqb_bhardlimit = src->d_blk_hardlimit;
+	dst->dqb_bsoftlimit = src->d_blk_softlimit;
+	dst->dqb_curspace = src->d_bcount;
+	dst->dqb_ihardlimit = src->d_ino_hardlimit;
+	dst->dqb_isoftlimit = src->d_ino_softlimit;
+	dst->dqb_curinodes = src->d_icount;
+	dst->dqb_btime = src->d_btimer;
+	dst->dqb_itime = src->d_itimer;
+	dst->dqb_valid = QIF_ALL;
+}
+
 static int quota_getquota(struct super_block *sb, int type, qid_t id,
 			  void __user *addr)
 {
+	struct fs_disk_quota fdq;
 	struct if_dqblk idq;
 	int ret;
 
-	if (!sb_has_quota_active(sb, type))
-		return -ESRCH;
 	if (!sb->s_qcop->get_dqblk)
 		return -ENOSYS;
-	ret = sb->s_qcop->get_dqblk(sb, type, id, &idq);
+	ret = sb->s_qcop->get_dqblk(sb, type, id, &fdq);
 	if (ret)
 		return ret;
+	copy_to_if_dqblk(&idq, &fdq);
 	if (copy_to_user(addr, &idq, sizeof(idq)))
 		return -EFAULT;
 	return 0;
 }
 
+static void copy_from_if_dqblk(struct fs_disk_quota *dst, struct if_dqblk *src)
+{
+	dst->d_blk_hardlimit = src->dqb_bhardlimit;
+	dst->d_blk_softlimit  = src->dqb_bsoftlimit;
+	dst->d_bcount = src->dqb_curspace;
+	dst->d_ino_hardlimit = src->dqb_ihardlimit;
+	dst->d_ino_softlimit = src->dqb_isoftlimit;
+	dst->d_icount = src->dqb_curinodes;
+	dst->d_btimer = src->dqb_btime;
+	dst->d_itimer = src->dqb_itime;
+
+	dst->d_fieldmask = 0;
+	if (src->dqb_valid & QIF_BLIMITS)
+		dst->d_fieldmask |= FS_DQ_BSOFT | FS_DQ_BHARD;
+	if (src->dqb_valid & QIF_SPACE)
+		dst->d_fieldmask |= FS_DQ_BCOUNT;
+	if (src->dqb_valid & QIF_ILIMITS)
+		dst->d_fieldmask |= FS_DQ_ISOFT | FS_DQ_IHARD;
+	if (src->dqb_valid & QIF_INODES)
+		dst->d_fieldmask |= FS_DQ_ICOUNT;
+	if (src->dqb_valid & QIF_BTIME)
+		dst->d_fieldmask |= FS_DQ_BTIMER;
+	if (src->dqb_valid & QIF_ITIME)
+		dst->d_fieldmask |= FS_DQ_ITIMER;
+}
+
 static int quota_setquota(struct super_block *sb, int type, qid_t id,
 			  void __user *addr)
 {
+	struct fs_disk_quota fdq;
 	struct if_dqblk idq;
 
 	if (copy_from_user(&idq, addr, sizeof(idq)))
 		return -EFAULT;
-	if (!sb_has_quota_active(sb, type))
-		return -ESRCH;
 	if (!sb->s_qcop->set_dqblk)
 		return -ENOSYS;
-	return sb->s_qcop->set_dqblk(sb, type, id, &idq);
+	copy_from_if_dqblk(&fdq, &idq);
+	return sb->s_qcop->set_dqblk(sb, type, id, &fdq);
 }
 
 static int quota_setxstate(struct super_block *sb, int cmd, void __user *addr)
@@ -199,9 +234,9 @@ static int quota_setxquota(struct super_block *sb, int type, qid_t id,
 
 	if (copy_from_user(&fdq, addr, sizeof(fdq)))
 		return -EFAULT;
-	if (!sb->s_qcop->set_xquota)
+	if (!sb->s_qcop->set_dqblk)
 		return -ENOSYS;
-	return sb->s_qcop->set_xquota(sb, type, id, &fdq);
+	return sb->s_qcop->set_dqblk(sb, type, id, &fdq);
 }
 
 static int quota_getxquota(struct super_block *sb, int type, qid_t id,
@@ -210,9 +245,9 @@ static int quota_getxquota(struct super_block *sb, int type, qid_t id,
 	struct fs_disk_quota fdq;
 	int ret;
 
-	if (!sb->s_qcop->get_xquota)
+	if (!sb->s_qcop->get_dqblk)
 		return -ENOSYS;
-	ret = sb->s_qcop->get_xquota(sb, type, id, &fdq);
+	ret = sb->s_qcop->get_dqblk(sb, type, id, &fdq);
 	if (!ret && copy_to_user(addr, &fdq, sizeof(fdq)))
 		return -EFAULT;
 	return ret;

fs/quota/quota_tree.c

@@ -60,9 +60,17 @@ static ssize_t read_blk(struct qtree_mem_dqinfo *info, uint blk, char *buf)
 static ssize_t write_blk(struct qtree_mem_dqinfo *info, uint blk, char *buf)
 {
 	struct super_block *sb = info->dqi_sb;
+	ssize_t ret;
 
-	return sb->s_op->quota_write(sb, info->dqi_type, buf,
+	ret = sb->s_op->quota_write(sb, info->dqi_type, buf,
 	       info->dqi_usable_bs, blk << info->dqi_blocksize_bits);
+	if (ret != info->dqi_usable_bs) {
+		q_warn(KERN_WARNING "VFS: dquota write failed on "
+			"dev %s\n", sb->s_id);
+		if (ret >= 0)
+			ret = -EIO;
+	}
+	return ret;
 }
 
 /* Remove empty block from list and return it */
@@ -152,7 +160,7 @@ static int remove_free_dqentry(struct qtree_mem_dqinfo *info, char *buf,
 	dh->dqdh_next_free = dh->dqdh_prev_free = cpu_to_le32(0);
 	/* No matter whether write succeeds block is out of list */
 	if (write_blk(info, blk, buf) < 0)
-		printk(KERN_ERR
+		q_warn(KERN_ERR
 		       "VFS: Can't write block (%u) with free entries.\n",
 		       blk);
 	return 0;
@@ -244,7 +252,7 @@ static uint find_free_dqentry(struct qtree_mem_dqinfo *info,
 	if (le16_to_cpu(dh->dqdh_entries) + 1 >= qtree_dqstr_in_blk(info)) {
 		*err = remove_free_dqentry(info, buf, blk);
 		if (*err < 0) {
-			printk(KERN_ERR "VFS: find_free_dqentry(): Can't "
+			q_warn(KERN_ERR "VFS: find_free_dqentry(): Can't "
 			       "remove block (%u) from entry free list.\n",
 			       blk);
 			goto out_buf;
@@ -268,7 +276,7 @@ static uint find_free_dqentry(struct qtree_mem_dqinfo *info,
 #endif
 	*err = write_blk(info, blk, buf);
 	if (*err < 0) {
-		printk(KERN_ERR "VFS: find_free_dqentry(): Can't write quota "
+		q_warn(KERN_ERR "VFS: find_free_dqentry(): Can't write quota "
 				"data block %u.\n", blk);
 		goto out_buf;
 	}
@@ -303,7 +311,7 @@ static int do_insert_tree(struct qtree_mem_dqinfo *info, struct dquot *dquot,
 	} else {
 		ret = read_blk(info, *treeblk, buf);
 		if (ret < 0) {
-			printk(KERN_ERR "VFS: Can't read tree quota block "
+			q_warn(KERN_ERR "VFS: Can't read tree quota block "
 					"%u.\n", *treeblk);
 			goto out_buf;
 		}
@@ -365,7 +373,7 @@ int qtree_write_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
 	if (!dquot->dq_off) {
 		ret = dq_insert_tree(info, dquot);
 		if (ret < 0) {
-			printk(KERN_ERR "VFS: Error %zd occurred while "
+			q_warn(KERN_ERR "VFS: Error %zd occurred while "
 					"creating quota.\n", ret);
 			kfree(ddquot);
 			return ret;
@@ -377,14 +385,14 @@ int qtree_write_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
 	ret = sb->s_op->quota_write(sb, type, ddquot, info->dqi_entry_size,
 				    dquot->dq_off);
 	if (ret != info->dqi_entry_size) {
-		printk(KERN_WARNING "VFS: dquota write failed on dev %s\n",
+		q_warn(KERN_WARNING "VFS: dquota write failed on dev %s\n",
 		       sb->s_id);
 		if (ret >= 0)
 			ret = -ENOSPC;
 	} else {
 		ret = 0;
 	}
-	dqstats.writes++;
+	dqstats_inc(DQST_WRITES);
 	kfree(ddquot);
 
 	return ret;
@@ -402,14 +410,14 @@ static int free_dqentry(struct qtree_mem_dqinfo *info, struct dquot *dquot,
 	if (!buf)
 		return -ENOMEM;
 	if (dquot->dq_off >> info->dqi_blocksize_bits != blk) {
-		printk(KERN_ERR "VFS: Quota structure has offset to other "
+		q_warn(KERN_ERR "VFS: Quota structure has offset to other "
 		  "block (%u) than it should (%u).\n", blk,
 		  (uint)(dquot->dq_off >> info->dqi_blocksize_bits));
 		goto out_buf;
 	}
 	ret = read_blk(info, blk, buf);
 	if (ret < 0) {
-		printk(KERN_ERR "VFS: Can't read quota data block %u\n", blk);
+		q_warn(KERN_ERR "VFS: Can't read quota data block %u\n", blk);
 		goto out_buf;
 	}
 	dh = (struct qt_disk_dqdbheader *)buf;
@@ -419,7 +427,7 @@ static int free_dqentry(struct qtree_mem_dqinfo *info, struct dquot *dquot,
 		if (ret >= 0)
 			ret = put_free_dqblk(info, buf, blk);
 		if (ret < 0) {
-			printk(KERN_ERR "VFS: Can't move quota data block (%u) "
+			q_warn(KERN_ERR "VFS: Can't move quota data block (%u) "
 			  "to free list.\n", blk);
 			goto out_buf;
 		}
@@ -432,14 +440,14 @@ static int free_dqentry(struct qtree_mem_dqinfo *info, struct dquot *dquot,
 			/* Insert will write block itself */
 			ret = insert_free_dqentry(info, buf, blk);
 			if (ret < 0) {
-				printk(KERN_ERR "VFS: Can't insert quota data "
+				q_warn(KERN_ERR "VFS: Can't insert quota data "
 				       "block (%u) to free entry list.\n", blk);
 				goto out_buf;
 			}
 		} else {
 			ret = write_blk(info, blk, buf);
 			if (ret < 0) {
-				printk(KERN_ERR "VFS: Can't write quota data "
+				q_warn(KERN_ERR "VFS: Can't write quota data "
 				  "block %u\n", blk);
 				goto out_buf;
 			}
@@ -464,7 +472,7 @@ static int remove_tree(struct qtree_mem_dqinfo *info, struct dquot *dquot,
 		return -ENOMEM;
 	ret = read_blk(info, *blk, buf);
 	if (ret < 0) {
-		printk(KERN_ERR "VFS: Can't read quota data block %u\n", *blk);
+		q_warn(KERN_ERR "VFS: Can't read quota data block %u\n", *blk);
 		goto out_buf;
 	}
 	newblk = le32_to_cpu(ref[get_index(info, dquot->dq_id, depth)]);
@@ -488,7 +496,7 @@ static int remove_tree(struct qtree_mem_dqinfo *info, struct dquot *dquot,
 		} else {
 			ret = write_blk(info, *blk, buf);
 			if (ret < 0)
-				printk(KERN_ERR "VFS: Can't write quota tree "
+				q_warn(KERN_ERR "VFS: Can't write quota tree "
 				  "block %u.\n", *blk);
 		}
 	}
@@ -521,7 +529,7 @@ static loff_t find_block_dqentry(struct qtree_mem_dqinfo *info,
 		return -ENOMEM;
 	ret = read_blk(info, blk, buf);
 	if (ret < 0) {
-		printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
+		q_warn(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
 		goto out_buf;
 	}
 	ddquot = buf + sizeof(struct qt_disk_dqdbheader);
@@ -531,7 +539,7 @@ static loff_t find_block_dqentry(struct qtree_mem_dqinfo *info,
 		ddquot += info->dqi_entry_size;
 	}
 	if (i == qtree_dqstr_in_blk(info)) {
-		printk(KERN_ERR "VFS: Quota for id %u referenced "
+		q_warn(KERN_ERR "VFS: Quota for id %u referenced "
 		  "but not present.\n", dquot->dq_id);
 		ret = -EIO;
 		goto out_buf;
@@ -556,7 +564,7 @@ static loff_t find_tree_dqentry(struct qtree_mem_dqinfo *info,
 		return -ENOMEM;
 	ret = read_blk(info, blk, buf);
 	if (ret < 0) {
-		printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
+		q_warn(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk);
 		goto out_buf;
 	}
 	ret = 0;
@@ -599,7 +607,7 @@ int qtree_read_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
 		offset = find_dqentry(info, dquot);
 		if (offset <= 0) {	/* Entry not present? */
 			if (offset < 0)
-				printk(KERN_ERR "VFS: Can't read quota "
+				q_warn(KERN_ERR "VFS: Can't read quota "
 				  "structure for id %u.\n", dquot->dq_id);
 			dquot->dq_off = 0;
 			set_bit(DQ_FAKE_B, &dquot->dq_flags);
@@ -617,7 +625,7 @@ int qtree_read_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
 	if (ret != info->dqi_entry_size) {
 		if (ret >= 0)
 			ret = -EIO;
-		printk(KERN_ERR "VFS: Error while reading quota "
+		q_warn(KERN_ERR "VFS: Error while reading quota "
 				"structure for id %u.\n", dquot->dq_id);
 		set_bit(DQ_FAKE_B, &dquot->dq_flags);
 		memset(&dquot->dq_dqb, 0, sizeof(struct mem_dqblk));
@@ -634,7 +642,7 @@ int qtree_read_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot)
 	spin_unlock(&dq_data_lock);
 	kfree(ddquot);
 out:
-	dqstats.reads++;
+	dqstats_inc(DQST_READS);
 	return ret;
 }
 EXPORT_SYMBOL(qtree_read_dquot);

fs/quota/quota_tree.h

@@ -22,4 +22,10 @@ struct qt_disk_dqdbheader {
 
 #define QT_TREEOFF	1		/* Offset of tree in file in blocks */
 
+#define q_warn(fmt, args...) \
+do { \
+	if (printk_ratelimit()) \
+		printk(fmt, ## args); \
+} while(0)
+
 #endif /* _LINUX_QUOTAIO_TREE_H */

fs/quota/quota_v1.c

@@ -71,7 +71,7 @@ static int v1_read_dqblk(struct dquot *dquot)
 	    dquot->dq_dqb.dqb_ihardlimit == 0 &&
 	    dquot->dq_dqb.dqb_isoftlimit == 0)
 		set_bit(DQ_FAKE_B, &dquot->dq_flags);
-	dqstats.reads++;
+	dqstats_inc(DQST_READS);
 
 	return 0;
 }
@@ -104,7 +104,7 @@ static int v1_commit_dqblk(struct dquot *dquot)
 	ret = 0;
 
 out:
-	dqstats.writes++;
+	dqstats_inc(DQST_WRITES);
 
 	return ret;
 }

fs/quota/quota_v2.c

@@ -63,7 +63,7 @@ static int v2_read_header(struct super_block *sb, int type,
 	size = sb->s_op->quota_read(sb, type, (char *)dqhead,
 				    sizeof(struct v2_disk_dqheader), 0);
 	if (size != sizeof(struct v2_disk_dqheader)) {
-		printk(KERN_WARNING "quota_v2: Failed header read:"
+		q_warn(KERN_WARNING "quota_v2: Failed header read:"
 		       " expected=%zd got=%zd\n",
 			sizeof(struct v2_disk_dqheader), size);
 		return 0;
@@ -106,7 +106,7 @@ static int v2_read_file_info(struct super_block *sb, int type)
 	size = sb->s_op->quota_read(sb, type, (char *)&dinfo,
 	       sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF);
 	if (size != sizeof(struct v2_disk_dqinfo)) {
-		printk(KERN_WARNING "quota_v2: Can't read info structure on device %s.\n",
+		q_warn(KERN_WARNING "quota_v2: Can't read info structure on device %s.\n",
 			sb->s_id);
 		return -1;
 	}
@@ -167,7 +167,7 @@ static int v2_write_file_info(struct super_block *sb, int type)
 	size = sb->s_op->quota_write(sb, type, (char *)&dinfo,
 	       sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF);
 	if (size != sizeof(struct v2_disk_dqinfo)) {
-		printk(KERN_WARNING "Can't write info structure on device %s.\n",
+		q_warn(KERN_WARNING "Can't write info structure on device %s.\n",
 			sb->s_id);
 		return -1;
 	}

fs/reiserfs/inode.c

@@ -3076,9 +3076,10 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 	ia_valid = attr->ia_valid &= ~(ATTR_KILL_SUID|ATTR_KILL_SGID);
 
 	depth = reiserfs_write_lock_once(inode->i_sb);
-	if (attr->ia_valid & ATTR_SIZE) {
+	if (is_quota_modification(inode, attr))
 		dquot_initialize(inode);
 
+	if (attr->ia_valid & ATTR_SIZE) {
 		/* version 2 items will be caught by the s_maxbytes check
 		 ** done for us in vmtruncate
 		 */

fs/udf/file.c

@@ -227,7 +227,7 @@ int udf_setattr(struct dentry *dentry, struct iattr *iattr)
 	if (error)
 		return error;
 
-	if (iattr->ia_valid & ATTR_SIZE)
+	if (is_quota_modification(inode, iattr))
 		dquot_initialize(inode);
 
 	if ((iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid) ||

fs/ufs/inode.c

@@ -603,7 +603,7 @@ static void ufs_set_inode_ops(struct inode *inode)
 		if (!inode->i_blocks)
 			inode->i_op = &ufs_fast_symlink_inode_operations;
 		else {
-			inode->i_op = &page_symlink_inode_operations;
+			inode->i_op = &ufs_symlink_inode_operations;
 			inode->i_mapping->a_ops = &ufs_aops;
 		}
 	} else

fs/ufs/namei.c

@@ -148,7 +148,7 @@ static int ufs_symlink (struct inode * dir, struct dentry * dentry,
 
 	if (l > UFS_SB(sb)->s_uspi->s_maxsymlinklen) {
 		/* slow symlink */
-		inode->i_op = &page_symlink_inode_operations;
+		inode->i_op = &ufs_symlink_inode_operations;
 		inode->i_mapping->a_ops = &ufs_aops;
 		err = page_symlink(inode, symname, l);
 		if (err)

fs/ufs/symlink.c

@@ -42,4 +42,12 @@ static void *ufs_follow_link(struct dentry *dentry, struct nameidata *nd)
 const struct inode_operations ufs_fast_symlink_inode_operations = {
 	.readlink	= generic_readlink,
 	.follow_link	= ufs_follow_link,
+	.setattr	= ufs_setattr,
+};
+
+const struct inode_operations ufs_symlink_inode_operations = {
+	.readlink	= generic_readlink,
+	.follow_link	= page_follow_link_light,
+	.put_link	= page_put_link,
+	.setattr	= ufs_setattr,
 };

fs/ufs/truncate.c

@@ -508,7 +508,7 @@ out:
  * - there is no way to know old size
  * - there is no way inform user about error, if it happens in `truncate'
  */
-static int ufs_setattr(struct dentry *dentry, struct iattr *attr)
+int ufs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
 	unsigned int ia_valid = attr->ia_valid;
@@ -518,18 +518,18 @@ static int ufs_setattr(struct dentry *dentry, struct iattr *attr)
 	if (error)
 		return error;
 
+	if (is_quota_modification(inode, attr))
+		dquot_initialize(inode);
+
 	if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
 	    (ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) {
 		error = dquot_transfer(inode, attr);
 		if (error)
 			return error;
 	}
-	if (ia_valid & ATTR_SIZE &&
-	    attr->ia_size != i_size_read(inode)) {
+	if (ia_valid & ATTR_SIZE && attr->ia_size != inode->i_size) {
 		loff_t old_i_size = inode->i_size;
 
-		dquot_initialize(inode);
-
 		error = vmtruncate(inode, attr->ia_size);
 		if (error)
 			return error;

fs/ufs/ufs.h

@@ -122,9 +122,11 @@ extern void ufs_panic (struct super_block *, const char *, const char *, ...) __
 
 /* symlink.c */
 extern const struct inode_operations ufs_fast_symlink_inode_operations;
+extern const struct inode_operations ufs_symlink_inode_operations;
 
 /* truncate.c */
 extern int ufs_truncate (struct inode *, loff_t);
+extern int ufs_setattr(struct dentry *dentry, struct iattr *attr);
 
 static inline struct ufs_sb_info *UFS_SB(struct super_block *sb)
 {

fs/xfs/linux-2.6/xfs_quotaops.c

@@ -97,7 +97,7 @@ xfs_fs_set_xstate(
 }
 
 STATIC int
-xfs_fs_get_xquota(
+xfs_fs_get_dqblk(
 	struct super_block	*sb,
 	int			type,
 	qid_t			id,
@@ -114,7 +114,7 @@ xfs_fs_get_xquota(
 }
 
 STATIC int
-xfs_fs_set_xquota(
+xfs_fs_set_dqblk(
 	struct super_block	*sb,
 	int			type,
 	qid_t			id,
@@ -135,6 +135,6 @@ xfs_fs_set_xquota(
 const struct quotactl_ops xfs_quotactl_operations = {
 	.get_xstate		= xfs_fs_get_xstate,
 	.set_xstate		= xfs_fs_set_xstate,
-	.get_xquota		= xfs_fs_get_xquota,
-	.set_xquota		= xfs_fs_set_xquota,
+	.get_dqblk		= xfs_fs_get_dqblk,
+	.set_dqblk		= xfs_fs_set_dqblk,
 };

fs/xfs/quota/xfs_qm_syscalls.c

@@ -448,6 +448,9 @@ xfs_qm_scall_getqstat(
 	return 0;
 }
 
+#define XFS_DQ_MASK \
+	(FS_DQ_LIMIT_MASK | FS_DQ_TIMER_MASK | FS_DQ_WARNS_MASK)
+
 /*
  * Adjust quota limits, and start/stop timers accordingly.
  */
@@ -465,9 +468,10 @@ xfs_qm_scall_setqlim(
 	int			error;
 	xfs_qcnt_t		hard, soft;
 
-	if ((newlim->d_fieldmask &
-	    (FS_DQ_LIMIT_MASK|FS_DQ_TIMER_MASK|FS_DQ_WARNS_MASK)) == 0)
-		return (0);
+	if (newlim->d_fieldmask & ~XFS_DQ_MASK)
+		return EINVAL;
+	if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0)
+		return 0;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM);
 	if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_disk_dquot_t) + 128,

include/linux/dqblk_xfs.h

@@ -109,6 +109,15 @@ typedef struct fs_disk_quota {
 #define FS_DQ_RTBWARNS	(1<<11)
 #define FS_DQ_WARNS_MASK	(FS_DQ_BWARNS | FS_DQ_IWARNS | FS_DQ_RTBWARNS)
 
+/*
+ * Accounting values.  These can only be set for filesystem with
+ * non-transactional quotas that require quotacheck(8) in userspace.
+ */
+#define FS_DQ_BCOUNT		(1<<12)
+#define FS_DQ_ICOUNT		(1<<13)
+#define FS_DQ_RTBCOUNT		(1<<14)
+#define FS_DQ_ACCT_MASK		(FS_DQ_BCOUNT | FS_DQ_ICOUNT | FS_DQ_RTBCOUNT)
+
 /*
  * Various flags related to quotactl(2).  Only relevant to XFS filesystems.
  */

include/linux/ext2_fs_sb.h

@@ -106,6 +106,15 @@ struct ext2_sb_info {
 	spinlock_t s_rsv_window_lock;
 	struct rb_root s_rsv_window_root;
 	struct ext2_reserve_window_node s_rsv_window_head;
+	/*
+	 * s_lock protects against concurrent modifications of s_mount_state,
+	 * s_blocks_last, s_overhead_last and the content of superblock's
+	 * buffer pointed to by sbi->s_es.
+	 *
+	 * Note: It is used in ext2_show_options() to provide a consistent view
+	 * of the mount options.
+	 */
+	spinlock_t s_lock;
 };
 
 static inline spinlock_t *

include/linux/jbd.h

@@ -427,9 +427,9 @@ struct transaction_s
 	enum {
 		T_RUNNING,
 		T_LOCKED,
-		T_RUNDOWN,
 		T_FLUSH,
 		T_COMMIT,
+		T_COMMIT_RECORD,
 		T_FINISHED
 	}			t_state;
 
@@ -991,6 +991,7 @@ int journal_start_commit(journal_t *journal, tid_t *tid);
 int journal_force_commit_nested(journal_t *journal);
 int log_wait_commit(journal_t *journal, tid_t tid);
 int log_do_checkpoint(journal_t *journal);
+int journal_trans_will_send_data_barrier(journal_t *journal, tid_t tid);
 
 void __log_wait_for_space(journal_t *journal);
 extern void	__journal_drop_transaction(journal_t *, transaction_t *);

include/linux/quota.h

@@ -174,6 +174,8 @@ enum {
 #include <linux/rwsem.h>
 #include <linux/spinlock.h>
 #include <linux/wait.h>
+#include <linux/percpu.h>
+#include <linux/smp.h>
 
 #include <linux/dqblk_xfs.h>
 #include <linux/dqblk_v1.h>
@@ -238,19 +240,43 @@ static inline int info_dirty(struct mem_dqinfo *info)
 	return test_bit(DQF_INFO_DIRTY_B, &info->dqi_flags);
 }
 
+enum {
+	DQST_LOOKUPS,
+	DQST_DROPS,
+	DQST_READS,
+	DQST_WRITES,
+	DQST_CACHE_HITS,
+	DQST_ALLOC_DQUOTS,
+	DQST_FREE_DQUOTS,
+	DQST_SYNCS,
+	_DQST_DQSTAT_LAST
+};
+
 struct dqstats {
-	int lookups;
-	int drops;
-	int reads;
-	int writes;
-	int cache_hits;
-	int allocated_dquots;
-	int free_dquots;
-	int syncs;
+	int stat[_DQST_DQSTAT_LAST];
 };
 
+extern struct dqstats *dqstats_pcpu;
 extern struct dqstats dqstats;
 
+static inline void dqstats_inc(unsigned int type)
+{
+#ifdef CONFIG_SMP
+	per_cpu_ptr(dqstats_pcpu, smp_processor_id())->stat[type]++;
+#else
+	dqstats.stat[type]++;
+#endif
+}
+
+static inline void dqstats_dec(unsigned int type)
+{
+#ifdef CONFIG_SMP
+	per_cpu_ptr(dqstats_pcpu, smp_processor_id())->stat[type]--;
+#else
+	dqstats.stat[type]--;
+#endif
+}
+
 #define DQ_MOD_B	0	/* dquot modified since read */
 #define DQ_BLKS_B	1	/* uid/gid has been warned about blk limit */
 #define DQ_INODES_B	2	/* uid/gid has been warned about inode limit */
@@ -311,12 +337,10 @@ struct quotactl_ops {
 	int (*quota_sync)(struct super_block *, int, int);
 	int (*get_info)(struct super_block *, int, struct if_dqinfo *);
 	int (*set_info)(struct super_block *, int, struct if_dqinfo *);
-	int (*get_dqblk)(struct super_block *, int, qid_t, struct if_dqblk *);
-	int (*set_dqblk)(struct super_block *, int, qid_t, struct if_dqblk *);
+	int (*get_dqblk)(struct super_block *, int, qid_t, struct fs_disk_quota *);
+	int (*set_dqblk)(struct super_block *, int, qid_t, struct fs_disk_quota *);
 	int (*get_xstate)(struct super_block *, struct fs_quota_stat *);
 	int (*set_xstate)(struct super_block *, unsigned int, int);
-	int (*get_xquota)(struct super_block *, int, qid_t, struct fs_disk_quota *);
-	int (*set_xquota)(struct super_block *, int, qid_t, struct fs_disk_quota *);
 };
 
 struct quota_format_type {

include/linux/quotaops.h

@@ -14,6 +14,14 @@ static inline struct quota_info *sb_dqopt(struct super_block *sb)
 	return &sb->s_dquot;
 }
 
+/* i_mutex must being held */
+static inline bool is_quota_modification(struct inode *inode, struct iattr *ia)
+{
+	return (ia->ia_valid & ATTR_SIZE && ia->ia_size != inode->i_size) ||
+		(ia->ia_valid & ATTR_UID && ia->ia_uid != inode->i_uid) ||
+		(ia->ia_valid & ATTR_GID && ia->ia_gid != inode->i_gid);
+}
+
 #if defined(CONFIG_QUOTA)
 
 /*
@@ -63,9 +71,12 @@ int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags);
 int vfs_quota_sync(struct super_block *sb, int type, int wait);
 int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
 int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
-int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
-int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
+int vfs_get_dqblk(struct super_block *sb, int type, qid_t id,
+		struct fs_disk_quota *di);
+int vfs_set_dqblk(struct super_block *sb, int type, qid_t id,
+		struct fs_disk_quota *di);
 
+int __dquot_transfer(struct inode *inode, struct dquot **transfer_to);
 int dquot_transfer(struct inode *inode, struct iattr *iattr);
 int vfs_dq_quota_on_remount(struct super_block *sb);