Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: remove write-only variables from ext4_ordered_write_end
  ext4: unexport jbd2_journal_update_superblock
  ext4: Cleanup whitespace and other miscellaneous style issues
  ext4: improve ext4_fill_flex_info() a bit
  ext4: Cleanup the block reservation code path
  ext4: don't assume extents can't cross block groups when truncating
  ext4: Fix lack of credits BUG() when deleting a badly fragmented inode
  ext4: Fix ext4_ext_journal_restart()
  ext4: fix ext4_da_write_begin error path
  jbd2: don't abort if flushing file data failed
  ext4: don't read inode block if the buffer has a write error
  ext4: Don't allow lg prealloc list to be grow large.
  ext4: Convert the usage of NR_CPUS to nr_cpu_ids.
  ext4: Improve error handling in mballoc
  ext4: lock block groups when initializing
  ext4: sync up block and inode bitmap reading functions
  ext4: Allow read/only mounts with corrupted block group checksums
  ext4: Fix data corruption when writing to prealloc area

fs/ext4/acl.c

@@ -40,34 +40,35 @@ ext4_acl_from_disk(const void *value, size_t size)
 	acl = posix_acl_alloc(count, GFP_NOFS);
 	if (!acl)
 		return ERR_PTR(-ENOMEM);
-	for (n=0; n < count; n++) {
+	for (n = 0; n < count; n++) {
 		ext4_acl_entry *entry =
 			(ext4_acl_entry *)value;
 		if ((char *)value + sizeof(ext4_acl_entry_short) > end)
 			goto fail;
 		acl->a_entries[n].e_tag  = le16_to_cpu(entry->e_tag);
 		acl->a_entries[n].e_perm = le16_to_cpu(entry->e_perm);
-		switch(acl->a_entries[n].e_tag) {
-			case ACL_USER_OBJ:
-			case ACL_GROUP_OBJ:
-			case ACL_MASK:
-			case ACL_OTHER:
-				value = (char *)value +
-					sizeof(ext4_acl_entry_short);
-				acl->a_entries[n].e_id = ACL_UNDEFINED_ID;
-				break;
-
-			case ACL_USER:
-			case ACL_GROUP:
-				value = (char *)value + sizeof(ext4_acl_entry);
-				if ((char *)value > end)
-					goto fail;
-				acl->a_entries[n].e_id =
-					le32_to_cpu(entry->e_id);
-				break;
-
-			default:
+
+		switch (acl->a_entries[n].e_tag) {
+		case ACL_USER_OBJ:
+		case ACL_GROUP_OBJ:
+		case ACL_MASK:
+		case ACL_OTHER:
+			value = (char *)value +
+				sizeof(ext4_acl_entry_short);
+			acl->a_entries[n].e_id = ACL_UNDEFINED_ID;
+			break;
+
+		case ACL_USER:
+		case ACL_GROUP:
+			value = (char *)value + sizeof(ext4_acl_entry);
+			if ((char *)value > end)
 				goto fail;
+			acl->a_entries[n].e_id =
+				le32_to_cpu(entry->e_id);
+			break;
+
+		default:
+			goto fail;
 		}
 	}
 	if (value != end)
@@ -96,27 +97,26 @@ ext4_acl_to_disk(const struct posix_acl *acl, size_t *size)
 		return ERR_PTR(-ENOMEM);
 	ext_acl->a_version = cpu_to_le32(EXT4_ACL_VERSION);
 	e = (char *)ext_acl + sizeof(ext4_acl_header);
-	for (n=0; n < acl->a_count; n++) {
+	for (n = 0; n < acl->a_count; n++) {
 		ext4_acl_entry *entry = (ext4_acl_entry *)e;
 		entry->e_tag  = cpu_to_le16(acl->a_entries[n].e_tag);
 		entry->e_perm = cpu_to_le16(acl->a_entries[n].e_perm);
-		switch(acl->a_entries[n].e_tag) {
-			case ACL_USER:
-			case ACL_GROUP:
-				entry->e_id =
-					cpu_to_le32(acl->a_entries[n].e_id);
-				e += sizeof(ext4_acl_entry);
-				break;
-
-			case ACL_USER_OBJ:
-			case ACL_GROUP_OBJ:
-			case ACL_MASK:
-			case ACL_OTHER:
-				e += sizeof(ext4_acl_entry_short);
-				break;
-
-			default:
-				goto fail;
+		switch (acl->a_entries[n].e_tag) {
+		case ACL_USER:
+		case ACL_GROUP:
+			entry->e_id = cpu_to_le32(acl->a_entries[n].e_id);
+			e += sizeof(ext4_acl_entry);
+			break;
+
+		case ACL_USER_OBJ:
+		case ACL_GROUP_OBJ:
+		case ACL_MASK:
+		case ACL_OTHER:
+			e += sizeof(ext4_acl_entry_short);
+			break;
+
+		default:
+			goto fail;
 		}
 	}
 	return (char *)ext_acl;
@@ -167,23 +167,23 @@ ext4_get_acl(struct inode *inode, int type)
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return NULL;
 
-	switch(type) {
-		case ACL_TYPE_ACCESS:
-			acl = ext4_iget_acl(inode, &ei->i_acl);
-			if (acl != EXT4_ACL_NOT_CACHED)
-				return acl;
-			name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS;
-			break;
-
-		case ACL_TYPE_DEFAULT:
-			acl = ext4_iget_acl(inode, &ei->i_default_acl);
-			if (acl != EXT4_ACL_NOT_CACHED)
-				return acl;
-			name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT;
-			break;
-
-		default:
-			return ERR_PTR(-EINVAL);
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		acl = ext4_iget_acl(inode, &ei->i_acl);
+		if (acl != EXT4_ACL_NOT_CACHED)
+			return acl;
+		name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS;
+		break;
+
+	case ACL_TYPE_DEFAULT:
+		acl = ext4_iget_acl(inode, &ei->i_default_acl);
+		if (acl != EXT4_ACL_NOT_CACHED)
+			return acl;
+		name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT;
+		break;
+
+	default:
+		return ERR_PTR(-EINVAL);
 	}
 	retval = ext4_xattr_get(inode, name_index, "", NULL, 0);
 	if (retval > 0) {
@@ -201,14 +201,14 @@ ext4_get_acl(struct inode *inode, int type)
 	kfree(value);
 
 	if (!IS_ERR(acl)) {
-		switch(type) {
-			case ACL_TYPE_ACCESS:
-				ext4_iset_acl(inode, &ei->i_acl, acl);
-				break;
-
-			case ACL_TYPE_DEFAULT:
-				ext4_iset_acl(inode, &ei->i_default_acl, acl);
-				break;
+		switch (type) {
+		case ACL_TYPE_ACCESS:
+			ext4_iset_acl(inode, &ei->i_acl, acl);
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			ext4_iset_acl(inode, &ei->i_default_acl, acl);
+			break;
 		}
 	}
 	return acl;
@@ -232,31 +232,31 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type,
 	if (S_ISLNK(inode->i_mode))
 		return -EOPNOTSUPP;
 
-	switch(type) {
-		case ACL_TYPE_ACCESS:
-			name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS;
-			if (acl) {
-				mode_t mode = inode->i_mode;
-				error = posix_acl_equiv_mode(acl, &mode);
-				if (error < 0)
-					return error;
-				else {
-					inode->i_mode = mode;
-					ext4_mark_inode_dirty(handle, inode);
-					if (error == 0)
-						acl = NULL;
-				}
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS;
+		if (acl) {
+			mode_t mode = inode->i_mode;
+			error = posix_acl_equiv_mode(acl, &mode);
+			if (error < 0)
+				return error;
+			else {
+				inode->i_mode = mode;
+				ext4_mark_inode_dirty(handle, inode);
+				if (error == 0)
+					acl = NULL;
 			}
-			break;
+		}
+		break;
 
-		case ACL_TYPE_DEFAULT:
-			name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT;
-			if (!S_ISDIR(inode->i_mode))
-				return acl ? -EACCES : 0;
-			break;
+	case ACL_TYPE_DEFAULT:
+		name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT;
+		if (!S_ISDIR(inode->i_mode))
+			return acl ? -EACCES : 0;
+		break;
 
-		default:
-			return -EINVAL;
+	default:
+		return -EINVAL;
 	}
 	if (acl) {
 		value = ext4_acl_to_disk(acl, &size);
@@ -269,14 +269,14 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type,
 
 	kfree(value);
 	if (!error) {
-		switch(type) {
-			case ACL_TYPE_ACCESS:
-				ext4_iset_acl(inode, &ei->i_acl, acl);
-				break;
-
-			case ACL_TYPE_DEFAULT:
-				ext4_iset_acl(inode, &ei->i_default_acl, acl);
-				break;
+		switch (type) {
+		case ACL_TYPE_ACCESS:
+			ext4_iset_acl(inode, &ei->i_acl, acl);
+			break;
+
+		case ACL_TYPE_DEFAULT:
+			ext4_iset_acl(inode, &ei->i_default_acl, acl);
+			break;
 		}
 	}
 	return error;

fs/ext4/balloc.c

@@ -314,25 +314,28 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 	if (unlikely(!bh)) {
 		ext4_error(sb, __func__,
 			    "Cannot read block bitmap - "
-			    "block_group = %d, block_bitmap = %llu",
-			    (int)block_group, (unsigned long long)bitmap_blk);
+			    "block_group = %lu, block_bitmap = %llu",
+			    block_group, bitmap_blk);
 		return NULL;
 	}
 	if (bh_uptodate_or_lock(bh))
 		return bh;
 
+	spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
 	if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
 		ext4_init_block_bitmap(sb, bh, block_group, desc);
 		set_buffer_uptodate(bh);
 		unlock_buffer(bh);
+		spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
 		return bh;
 	}
+	spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
 	if (bh_submit_read(bh) < 0) {
 		put_bh(bh);
 		ext4_error(sb, __func__,
 			    "Cannot read block bitmap - "
-			    "block_group = %d, block_bitmap = %llu",
-			    (int)block_group, (unsigned long long)bitmap_blk);
+			    "block_group = %lu, block_bitmap = %llu",
+			    block_group, bitmap_blk);
 		return NULL;
 	}
 	ext4_valid_block_bitmap(sb, desc, block_group, bh);

fs/ext4/ext4.h

@@ -1044,7 +1044,6 @@ extern void ext4_mb_update_group_info(struct ext4_group_info *grp,
 
 
 /* inode.c */
-void ext4_da_release_space(struct inode *inode, int used, int to_free);
 int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
 		struct buffer_head *bh, ext4_fsblk_t blocknr);
 struct buffer_head *ext4_getblk(handle_t *, struct inode *,

fs/ext4/extents.c

@@ -99,7 +99,7 @@ static int ext4_ext_journal_restart(handle_t *handle, int needed)
 	if (handle->h_buffer_credits > needed)
 		return 0;
 	err = ext4_journal_extend(handle, needed);
-	if (err)
+	if (err <= 0)
 		return err;
 	return ext4_journal_restart(handle, needed);
 }
@@ -1441,7 +1441,7 @@ unsigned int ext4_ext_check_overlap(struct inode *inode,
 
 	/*
 	 * get the next allocated block if the extent in the path
-	 * is before the requested block(s) 
+	 * is before the requested block(s)
 	 */
 	if (b2 < b1) {
 		b2 = ext4_ext_next_allocated_block(path);
@@ -1910,9 +1910,13 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
 			BUG_ON(b != ex_ee_block + ex_ee_len - 1);
 		}
 
-		/* at present, extent can't cross block group: */
-		/* leaf + bitmap + group desc + sb + inode */
-		credits = 5;
+		/*
+		 * 3 for leaf, sb, and inode plus 2 (bmap and group
+		 * descriptor) for each block group; assume two block
+		 * groups plus ex_ee_len/blocks_per_block_group for
+		 * the worst case
+		 */
+		credits = 7 + 2*(ex_ee_len/EXT4_BLOCKS_PER_GROUP(inode->i_sb));
 		if (ex == EXT_FIRST_EXTENT(eh)) {
 			correct_index = 1;
 			credits += (ext_depth(inode)) + 1;
@@ -2323,7 +2327,10 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 		unsigned int newdepth;
 		/* If extent has less than EXT4_EXT_ZERO_LEN zerout directly */
 		if (allocated <= EXT4_EXT_ZERO_LEN) {
-			/* Mark first half uninitialized.
+			/*
+			 * iblock == ee_block is handled by the zerouout
+			 * at the beginning.
+			 * Mark first half uninitialized.
 			 * Mark second half initialized and zero out the
 			 * initialized extent
 			 */
@@ -2346,7 +2353,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 				ex->ee_len   = orig_ex.ee_len;
 				ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
 				ext4_ext_dirty(handle, inode, path + depth);
-				/* zeroed the full extent */
+				/* blocks available from iblock */
 				return allocated;
 
 			} else if (err)
@@ -2374,6 +2381,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 					err = PTR_ERR(path);
 					return err;
 				}
+				/* get the second half extent details */
 				ex = path[depth].p_ext;
 				err = ext4_ext_get_access(handle, inode,
 								path + depth);
@@ -2403,6 +2411,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 			ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
 			ext4_ext_dirty(handle, inode, path + depth);
 			/* zeroed the full extent */
+			/* blocks available from iblock */
 			return allocated;
 
 		} else if (err)
@@ -2418,23 +2427,22 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 		 */
 		orig_ex.ee_len = cpu_to_le16(ee_len -
 						ext4_ext_get_actual_len(ex3));
-		if (newdepth != depth) {
-			depth = newdepth;
-			ext4_ext_drop_refs(path);
-			path = ext4_ext_find_extent(inode, iblock, path);
-			if (IS_ERR(path)) {
-				err = PTR_ERR(path);
-				goto out;
-			}
-			eh = path[depth].p_hdr;
-			ex = path[depth].p_ext;
-			if (ex2 != &newex)
-				ex2 = ex;
-
-			err = ext4_ext_get_access(handle, inode, path + depth);
-			if (err)
-				goto out;
+		depth = newdepth;
+		ext4_ext_drop_refs(path);
+		path = ext4_ext_find_extent(inode, iblock, path);
+		if (IS_ERR(path)) {
+			err = PTR_ERR(path);
+			goto out;
 		}
+		eh = path[depth].p_hdr;
+		ex = path[depth].p_ext;
+		if (ex2 != &newex)
+			ex2 = ex;
+
+		err = ext4_ext_get_access(handle, inode, path + depth);
+		if (err)
+			goto out;
+
 		allocated = max_blocks;
 
 		/* If extent has less than EXT4_EXT_ZERO_LEN and we are trying
@@ -2452,6 +2460,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 			ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
 			ext4_ext_dirty(handle, inode, path + depth);
 			/* zero out the first half */
+			/* blocks available from iblock */
 			return allocated;
 		}
 	}

fs/ext4/ialloc.c

@@ -97,34 +97,44 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh,
  * Return buffer_head of bitmap on success or NULL.
  */
 static struct buffer_head *
-read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
+ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
 {
 	struct ext4_group_desc *desc;
 	struct buffer_head *bh = NULL;
+	ext4_fsblk_t bitmap_blk;
 
 	desc = ext4_get_group_desc(sb, block_group, NULL);
 	if (!desc)
-		goto error_out;
-	if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
-		bh = sb_getblk(sb, ext4_inode_bitmap(sb, desc));
-		if (!buffer_uptodate(bh)) {
-			lock_buffer(bh);
-			if (!buffer_uptodate(bh)) {
-				ext4_init_inode_bitmap(sb, bh, block_group,
-						       desc);
-				set_buffer_uptodate(bh);
-			}
-			unlock_buffer(bh);
-		}
-	} else {
-		bh = sb_bread(sb, ext4_inode_bitmap(sb, desc));
+		return NULL;
+	bitmap_blk = ext4_inode_bitmap(sb, desc);
+	bh = sb_getblk(sb, bitmap_blk);
+	if (unlikely(!bh)) {
+		ext4_error(sb, __func__,
+			    "Cannot read inode bitmap - "
+			    "block_group = %lu, inode_bitmap = %llu",
+			    block_group, bitmap_blk);
+		return NULL;
 	}
-	if (!bh)
-		ext4_error(sb, "read_inode_bitmap",
+	if (bh_uptodate_or_lock(bh))
+		return bh;
+
+	spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
+	if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
+		ext4_init_inode_bitmap(sb, bh, block_group, desc);
+		set_buffer_uptodate(bh);
+		unlock_buffer(bh);
+		spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
+		return bh;
+	}
+	spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group));
+	if (bh_submit_read(bh) < 0) {
+		put_bh(bh);
+		ext4_error(sb, __func__,
 			    "Cannot read inode bitmap - "
 			    "block_group = %lu, inode_bitmap = %llu",
-			    block_group, ext4_inode_bitmap(sb, desc));
-error_out:
+			    block_group, bitmap_blk);
+		return NULL;
+	}
 	return bh;
 }
 
@@ -200,7 +210,7 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
 	}
 	block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
 	bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
-	bitmap_bh = read_inode_bitmap(sb, block_group);
+	bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
 	if (!bitmap_bh)
 		goto error_return;
 
@@ -623,7 +633,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
 			goto fail;
 
 		brelse(bitmap_bh);
-		bitmap_bh = read_inode_bitmap(sb, group);
+		bitmap_bh = ext4_read_inode_bitmap(sb, group);
 		if (!bitmap_bh)
 			goto fail;
 
@@ -728,7 +738,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
 
 			/* When marking the block group with
 			 * ~EXT4_BG_INODE_UNINIT we don't want to depend
-			 * on the value of bg_itable_unsed even though
+			 * on the value of bg_itable_unused even though
 			 * mke2fs could have initialized the same for us.
 			 * Instead we calculated the value below
 			 */
@@ -891,7 +901,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
 
 	block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
 	bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
-	bitmap_bh = read_inode_bitmap(sb, block_group);
+	bitmap_bh = ext4_read_inode_bitmap(sb, block_group);
 	if (!bitmap_bh) {
 		ext4_warning(sb, __func__,
 			     "inode bitmap error for orphan %lu", ino);
@@ -969,7 +979,7 @@ unsigned long ext4_count_free_inodes (struct super_block * sb)
 			continue;
 		desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
 		brelse(bitmap_bh);
-		bitmap_bh = read_inode_bitmap(sb, i);
+		bitmap_bh = ext4_read_inode_bitmap(sb, i);
 		if (!bitmap_bh)
 			continue;
 

fs/ext4/inode.c

@@ -191,6 +191,7 @@ static int ext4_journal_test_restart(handle_t *handle, struct inode *inode)
 void ext4_delete_inode (struct inode * inode)
 {
 	handle_t *handle;
+	int err;
 
 	if (ext4_should_order_data(inode))
 		ext4_begin_ordered_truncate(inode, 0);
@@ -199,8 +200,9 @@ void ext4_delete_inode (struct inode * inode)
 	if (is_bad_inode(inode))
 		goto no_delete;
 
-	handle = start_transaction(inode);
+	handle = ext4_journal_start(inode, blocks_for_truncate(inode)+3);
 	if (IS_ERR(handle)) {
+		ext4_std_error(inode->i_sb, PTR_ERR(handle));
 		/*
 		 * If we're going to skip the normal cleanup, we still need to
 		 * make sure that the in-core orphan linked list is properly
@@ -213,8 +215,34 @@ void ext4_delete_inode (struct inode * inode)
 	if (IS_SYNC(inode))
 		handle->h_sync = 1;
 	inode->i_size = 0;
+	err = ext4_mark_inode_dirty(handle, inode);
+	if (err) {
+		ext4_warning(inode->i_sb, __func__,
+			     "couldn't mark inode dirty (err %d)", err);
+		goto stop_handle;
+	}
 	if (inode->i_blocks)
 		ext4_truncate(inode);
+
+	/*
+	 * ext4_ext_truncate() doesn't reserve any slop when it
+	 * restarts journal transactions; therefore there may not be
+	 * enough credits left in the handle to remove the inode from
+	 * the orphan list and set the dtime field.
+	 */
+	if (handle->h_buffer_credits < 3) {
+		err = ext4_journal_extend(handle, 3);
+		if (err > 0)
+			err = ext4_journal_restart(handle, 3);
+		if (err != 0) {
+			ext4_warning(inode->i_sb, __func__,
+				     "couldn't extend journal (err %d)", err);
+		stop_handle:
+			ext4_journal_stop(handle);
+			goto no_delete;
+		}
+	}
+
 	/*
 	 * Kill off the orphan record which ext4_truncate created.
 	 * AKPM: I think this can be inside the above `if'.
@@ -952,6 +980,67 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 	return err;
 }
 
+/*
+ * Calculate the number of metadata blocks need to reserve
+ * to allocate @blocks for non extent file based file
+ */
+static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks)
+{
+	int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb);
+	int ind_blks, dind_blks, tind_blks;
+
+	/* number of new indirect blocks needed */
+	ind_blks = (blocks + icap - 1) / icap;
+
+	dind_blks = (ind_blks + icap - 1) / icap;
+
+	tind_blks = 1;
+
+	return ind_blks + dind_blks + tind_blks;
+}
+
+/*
+ * Calculate the number of metadata blocks need to reserve
+ * to allocate given number of blocks
+ */
+static int ext4_calc_metadata_amount(struct inode *inode, int blocks)
+{
+	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)
+		return ext4_ext_calc_metadata_amount(inode, blocks);
+
+	return ext4_indirect_calc_metadata_amount(inode, blocks);
+}
+
+static void ext4_da_update_reserve_space(struct inode *inode, int used)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+	int total, mdb, mdb_free;
+
+	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
+	/* recalculate the number of metablocks still need to be reserved */
+	total = EXT4_I(inode)->i_reserved_data_blocks - used;
+	mdb = ext4_calc_metadata_amount(inode, total);
+
+	/* figure out how many metablocks to release */
+	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
+	mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb;
+
+	/* Account for allocated meta_blocks */
+	mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks;
+
+	/* update fs free blocks counter for truncate case */
+	percpu_counter_add(&sbi->s_freeblocks_counter, mdb_free);
+
+	/* update per-inode reservations */
+	BUG_ON(used  > EXT4_I(inode)->i_reserved_data_blocks);
+	EXT4_I(inode)->i_reserved_data_blocks -= used;
+
+	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
+	EXT4_I(inode)->i_reserved_meta_blocks = mdb;
+	EXT4_I(inode)->i_allocated_meta_blocks = 0;
+	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
+}
+
 /* Maximum number of blocks we map for direct IO at once. */
 #define DIO_MAX_BLOCKS 4096
 /*
@@ -965,10 +1054,9 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 
 
 /*
+ * The ext4_get_blocks_wrap() function try to look up the requested blocks,
+ * and returns if the blocks are already mapped.
  *
- *
- * ext4_ext4 get_block() wrapper function
- * It will do a look up first, and returns if the blocks already mapped.
  * Otherwise it takes the write lock of the i_data_sem and allocate blocks
  * and store the allocated blocks in the result buffer head and mark it
  * mapped.
@@ -1069,7 +1157,7 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
 		 * which were deferred till now
 		 */
 		if ((retval > 0) && buffer_delay(bh))
-			ext4_da_release_space(inode, retval, 0);
+			ext4_da_update_reserve_space(inode, retval);
 	}
 
 	up_write((&EXT4_I(inode)->i_data_sem));
@@ -1336,12 +1424,8 @@ static int ext4_ordered_write_end(struct file *file,
 {
 	handle_t *handle = ext4_journal_current_handle();
 	struct inode *inode = mapping->host;
-	unsigned from, to;
 	int ret = 0, ret2;
 
-	from = pos & (PAGE_CACHE_SIZE - 1);
-	to = from + len;
-
 	ret = ext4_jbd2_file_inode(handle, inode);
 
 	if (ret == 0) {
@@ -1437,36 +1521,6 @@ static int ext4_journalled_write_end(struct file *file,
 
 	return ret ? ret : copied;
 }
-/*
- * Calculate the number of metadata blocks need to reserve
- * to allocate @blocks for non extent file based file
- */
-static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks)
-{
-	int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb);
-	int ind_blks, dind_blks, tind_blks;
-
-	/* number of new indirect blocks needed */
-	ind_blks = (blocks + icap - 1) / icap;
-
-	dind_blks = (ind_blks + icap - 1) / icap;
-
-	tind_blks = 1;
-
-	return ind_blks + dind_blks + tind_blks;
-}
-
-/*
- * Calculate the number of metadata blocks need to reserve
- * to allocate given number of blocks
- */
-static int ext4_calc_metadata_amount(struct inode *inode, int blocks)
-{
-	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)
-		return ext4_ext_calc_metadata_amount(inode, blocks);
-
-	return ext4_indirect_calc_metadata_amount(inode, blocks);
-}
 
 static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
 {
@@ -1490,7 +1544,6 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
 		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 		return -ENOSPC;
 	}
-
 	/* reduce fs free blocks counter */
 	percpu_counter_sub(&sbi->s_freeblocks_counter, total);
 
@@ -1501,35 +1554,31 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
 	return 0;       /* success */
 }
 
-void ext4_da_release_space(struct inode *inode, int used, int to_free)
+static void ext4_da_release_space(struct inode *inode, int to_free)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	int total, mdb, mdb_free, release;
 
 	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
 	/* recalculate the number of metablocks still need to be reserved */
-	total = EXT4_I(inode)->i_reserved_data_blocks - used - to_free;
+	total = EXT4_I(inode)->i_reserved_data_blocks - to_free;
 	mdb = ext4_calc_metadata_amount(inode, total);
 
 	/* figure out how many metablocks to release */
 	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
 	mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb;
 
-	/* Account for allocated meta_blocks */
-	mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks;
-
 	release = to_free + mdb_free;
 
 	/* update fs free blocks counter for truncate case */
 	percpu_counter_add(&sbi->s_freeblocks_counter, release);
 
 	/* update per-inode reservations */
-	BUG_ON(used + to_free > EXT4_I(inode)->i_reserved_data_blocks);
-	EXT4_I(inode)->i_reserved_data_blocks -= (used + to_free);
+	BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks);
+	EXT4_I(inode)->i_reserved_data_blocks -= to_free;
 
 	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
 	EXT4_I(inode)->i_reserved_meta_blocks = mdb;
-	EXT4_I(inode)->i_allocated_meta_blocks = 0;
 	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 }
 
@@ -1551,7 +1600,7 @@ static void ext4_da_page_release_reservation(struct page *page,
 		}
 		curr_off = next_off;
 	} while ((bh = bh->b_this_page) != head);
-	ext4_da_release_space(page->mapping->host, 0, to_release);
+	ext4_da_release_space(page->mapping->host, to_release);
 }
 
 /*
@@ -2280,8 +2329,11 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 	}
 
 	page = __grab_cache_page(mapping, index);
-	if (!page)
-		return -ENOMEM;
+	if (!page) {
+		ext4_journal_stop(handle);
+		ret = -ENOMEM;
+		goto out;
+	}
 	*pagep = page;
 
 	ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata,
@@ -3590,6 +3642,16 @@ static int __ext4_get_inode_loc(struct inode *inode,
 	}
 	if (!buffer_uptodate(bh)) {
 		lock_buffer(bh);
+
+		/*
+		 * If the buffer has the write error flag, we have failed
+		 * to write out another inode in the same block.  In this
+		 * case, we don't have to read the block because we may
+		 * read the old inode data successfully.
+		 */
+		if (buffer_write_io_error(bh) && !buffer_uptodate(bh))
+			set_buffer_uptodate(bh);
+
 		if (buffer_uptodate(bh)) {
 			/* someone brought it uptodate while we waited */
 			unlock_buffer(bh);

fs/ext4/mballoc.c

@@ -787,13 +787,16 @@ static int ext4_mb_init_cache(struct page *page, char *incore)
 		if (bh_uptodate_or_lock(bh[i]))
 			continue;
 
+		spin_lock(sb_bgl_lock(EXT4_SB(sb), first_group + i));
 		if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
 			ext4_init_block_bitmap(sb, bh[i],
 						first_group + i, desc);
 			set_buffer_uptodate(bh[i]);
 			unlock_buffer(bh[i]);
+			spin_unlock(sb_bgl_lock(EXT4_SB(sb), first_group + i));
 			continue;
 		}
+		spin_unlock(sb_bgl_lock(EXT4_SB(sb), first_group + i));
 		get_bh(bh[i]);
 		bh[i]->b_end_io = end_buffer_read_sync;
 		submit_bh(READ, bh[i]);
@@ -2477,7 +2480,7 @@ static int ext4_mb_init_backend(struct super_block *sb)
 int ext4_mb_init(struct super_block *sb, int needs_recovery)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	unsigned i;
+	unsigned i, j;
 	unsigned offset;
 	unsigned max;
 	int ret;
@@ -2537,7 +2540,7 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery)
 	sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT;
 	sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC;
 
-	i = sizeof(struct ext4_locality_group) * NR_CPUS;
+	i = sizeof(struct ext4_locality_group) * nr_cpu_ids;
 	sbi->s_locality_groups = kmalloc(i, GFP_KERNEL);
 	if (sbi->s_locality_groups == NULL) {
 		clear_opt(sbi->s_mount_opt, MBALLOC);
@@ -2545,11 +2548,12 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery)
 		kfree(sbi->s_mb_maxs);
 		return -ENOMEM;
 	}
-	for (i = 0; i < NR_CPUS; i++) {
+	for (i = 0; i < nr_cpu_ids; i++) {
 		struct ext4_locality_group *lg;
 		lg = &sbi->s_locality_groups[i];
 		mutex_init(&lg->lg_mutex);
-		INIT_LIST_HEAD(&lg->lg_prealloc_list);
+		for (j = 0; j < PREALLOC_TB_SIZE; j++)
+			INIT_LIST_HEAD(&lg->lg_prealloc_list[j]);
 		spin_lock_init(&lg->lg_prealloc_lock);
 	}
 
@@ -3260,6 +3264,7 @@ static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac,
 				struct ext4_prealloc_space *pa)
 {
 	unsigned int len = ac->ac_o_ex.fe_len;
+
 	ext4_get_group_no_and_offset(ac->ac_sb, pa->pa_pstart,
 					&ac->ac_b_ex.fe_group,
 					&ac->ac_b_ex.fe_start);
@@ -3282,6 +3287,7 @@ static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac,
 static noinline_for_stack int
 ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
 {
+	int order, i;
 	struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
 	struct ext4_locality_group *lg;
 	struct ext4_prealloc_space *pa;
@@ -3322,22 +3328,29 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
 	lg = ac->ac_lg;
 	if (lg == NULL)
 		return 0;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(pa, &lg->lg_prealloc_list, pa_inode_list) {
-		spin_lock(&pa->pa_lock);
-		if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) {
-			atomic_inc(&pa->pa_count);
-			ext4_mb_use_group_pa(ac, pa);
+	order  = fls(ac->ac_o_ex.fe_len) - 1;
+	if (order > PREALLOC_TB_SIZE - 1)
+		/* The max size of hash table is PREALLOC_TB_SIZE */
+		order = PREALLOC_TB_SIZE - 1;
+
+	for (i = order; i < PREALLOC_TB_SIZE; i++) {
+		rcu_read_lock();
+		list_for_each_entry_rcu(pa, &lg->lg_prealloc_list[i],
+					pa_inode_list) {
+			spin_lock(&pa->pa_lock);
+			if (pa->pa_deleted == 0 &&
+					pa->pa_free >= ac->ac_o_ex.fe_len) {
+				atomic_inc(&pa->pa_count);
+				ext4_mb_use_group_pa(ac, pa);
+				spin_unlock(&pa->pa_lock);
+				ac->ac_criteria = 20;
+				rcu_read_unlock();
+				return 1;
+			}
 			spin_unlock(&pa->pa_lock);
-			ac->ac_criteria = 20;
-			rcu_read_unlock();
-			return 1;
 		}
-		spin_unlock(&pa->pa_lock);
+		rcu_read_unlock();
 	}
-	rcu_read_unlock();
-
 	return 0;
 }
 
@@ -3560,6 +3573,7 @@ ext4_mb_new_group_pa(struct ext4_allocation_context *ac)
 	pa->pa_free = pa->pa_len;
 	atomic_set(&pa->pa_count, 1);
 	spin_lock_init(&pa->pa_lock);
+	INIT_LIST_HEAD(&pa->pa_inode_list);
 	pa->pa_deleted = 0;
 	pa->pa_linear = 1;
 
@@ -3580,10 +3594,10 @@ ext4_mb_new_group_pa(struct ext4_allocation_context *ac)
 	list_add(&pa->pa_group_list, &grp->bb_prealloc_list);
 	ext4_unlock_group(sb, ac->ac_b_ex.fe_group);
 
-	spin_lock(pa->pa_obj_lock);
-	list_add_tail_rcu(&pa->pa_inode_list, &lg->lg_prealloc_list);
-	spin_unlock(pa->pa_obj_lock);
-
+	/*
+	 * We will later add the new pa to the right bucket
+	 * after updating the pa_free in ext4_mb_release_context
+	 */
 	return 0;
 }
 
@@ -3733,20 +3747,23 @@ ext4_mb_discard_group_preallocations(struct super_block *sb,
 
 	bitmap_bh = ext4_read_block_bitmap(sb, group);
 	if (bitmap_bh == NULL) {
-		/* error handling here */
-		ext4_mb_release_desc(&e4b);
-		BUG_ON(bitmap_bh == NULL);
+		ext4_error(sb, __func__, "Error in reading block "
+				"bitmap for %lu\n", group);
+		return 0;
 	}
 
 	err = ext4_mb_load_buddy(sb, group, &e4b);
-	BUG_ON(err != 0); /* error handling here */
+	if (err) {
+		ext4_error(sb, __func__, "Error in loading buddy "
+				"information for %lu\n", group);
+		put_bh(bitmap_bh);
+		return 0;
+	}
 
 	if (needed == 0)
 		needed = EXT4_BLOCKS_PER_GROUP(sb) + 1;
 
-	grp = ext4_get_group_info(sb, group);
 	INIT_LIST_HEAD(&list);
-
 	ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
 repeat:
 	ext4_lock_group(sb, group);
@@ -3903,13 +3920,18 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode)
 		ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL);
 
 		err = ext4_mb_load_buddy(sb, group, &e4b);
-		BUG_ON(err != 0); /* error handling here */
+		if (err) {
+			ext4_error(sb, __func__, "Error in loading buddy "
+					"information for %lu\n", group);
+			continue;
+		}
 
 		bitmap_bh = ext4_read_block_bitmap(sb, group);
 		if (bitmap_bh == NULL) {
-			/* error handling here */
+			ext4_error(sb, __func__, "Error in reading block "
+					"bitmap for %lu\n", group);
 			ext4_mb_release_desc(&e4b);
-			BUG_ON(bitmap_bh == NULL);
+			continue;
 		}
 
 		ext4_lock_group(sb, group);
@@ -4112,22 +4134,168 @@ ext4_mb_initialize_context(struct ext4_allocation_context *ac,
 
 }
 
+static noinline_for_stack void
+ext4_mb_discard_lg_preallocations(struct super_block *sb,
+					struct ext4_locality_group *lg,
+					int order, int total_entries)
+{
+	ext4_group_t group = 0;
+	struct ext4_buddy e4b;
+	struct list_head discard_list;
+	struct ext4_prealloc_space *pa, *tmp;
+	struct ext4_allocation_context *ac;
+
+	mb_debug("discard locality group preallocation\n");
+
+	INIT_LIST_HEAD(&discard_list);
+	ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
+
+	spin_lock(&lg->lg_prealloc_lock);
+	list_for_each_entry_rcu(pa, &lg->lg_prealloc_list[order],
+						pa_inode_list) {
+		spin_lock(&pa->pa_lock);
+		if (atomic_read(&pa->pa_count)) {
+			/*
+			 * This is the pa that we just used
+			 * for block allocation. So don't
+			 * free that
+			 */
+			spin_unlock(&pa->pa_lock);
+			continue;
+		}
+		if (pa->pa_deleted) {
+			spin_unlock(&pa->pa_lock);
+			continue;
+		}
+		/* only lg prealloc space */
+		BUG_ON(!pa->pa_linear);
+
+		/* seems this one can be freed ... */
+		pa->pa_deleted = 1;
+		spin_unlock(&pa->pa_lock);
+
+		list_del_rcu(&pa->pa_inode_list);
+		list_add(&pa->u.pa_tmp_list, &discard_list);
+
+		total_entries--;
+		if (total_entries <= 5) {
+			/*
+			 * we want to keep only 5 entries
+			 * allowing it to grow to 8. This
+			 * mak sure we don't call discard
+			 * soon for this list.
+			 */
+			break;
+		}
+	}
+	spin_unlock(&lg->lg_prealloc_lock);
+
+	list_for_each_entry_safe(pa, tmp, &discard_list, u.pa_tmp_list) {
+
+		ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL);
+		if (ext4_mb_load_buddy(sb, group, &e4b)) {
+			ext4_error(sb, __func__, "Error in loading buddy "
+					"information for %lu\n", group);
+			continue;
+		}
+		ext4_lock_group(sb, group);
+		list_del(&pa->pa_group_list);
+		ext4_mb_release_group_pa(&e4b, pa, ac);
+		ext4_unlock_group(sb, group);
+
+		ext4_mb_release_desc(&e4b);
+		list_del(&pa->u.pa_tmp_list);
+		call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback);
+	}
+	if (ac)
+		kmem_cache_free(ext4_ac_cachep, ac);
+}
+
+/*
+ * We have incremented pa_count. So it cannot be freed at this
+ * point. Also we hold lg_mutex. So no parallel allocation is
+ * possible from this lg. That means pa_free cannot be updated.
+ *
+ * A parallel ext4_mb_discard_group_preallocations is possible.
+ * which can cause the lg_prealloc_list to be updated.
+ */
+
+static void ext4_mb_add_n_trim(struct ext4_allocation_context *ac)
+{
+	int order, added = 0, lg_prealloc_count = 1;
+	struct super_block *sb = ac->ac_sb;
+	struct ext4_locality_group *lg = ac->ac_lg;
+	struct ext4_prealloc_space *tmp_pa, *pa = ac->ac_pa;
+
+	order = fls(pa->pa_free) - 1;
+	if (order > PREALLOC_TB_SIZE - 1)
+		/* The max size of hash table is PREALLOC_TB_SIZE */
+		order = PREALLOC_TB_SIZE - 1;
+	/* Add the prealloc space to lg */
+	rcu_read_lock();
+	list_for_each_entry_rcu(tmp_pa, &lg->lg_prealloc_list[order],
+						pa_inode_list) {
+		spin_lock(&tmp_pa->pa_lock);
+		if (tmp_pa->pa_deleted) {
+			spin_unlock(&pa->pa_lock);
+			continue;
+		}
+		if (!added && pa->pa_free < tmp_pa->pa_free) {
+			/* Add to the tail of the previous entry */
+			list_add_tail_rcu(&pa->pa_inode_list,
+						&tmp_pa->pa_inode_list);
+			added = 1;
+			/*
+			 * we want to count the total
+			 * number of entries in the list
+			 */
+		}
+		spin_unlock(&tmp_pa->pa_lock);
+		lg_prealloc_count++;
+	}
+	if (!added)
+		list_add_tail_rcu(&pa->pa_inode_list,
+					&lg->lg_prealloc_list[order]);
+	rcu_read_unlock();
+
+	/* Now trim the list to be not more than 8 elements */
+	if (lg_prealloc_count > 8) {
+		ext4_mb_discard_lg_preallocations(sb, lg,
+						order, lg_prealloc_count);
+		return;
+	}
+	return ;
+}
+
 /*
  * release all resource we used in allocation
  */
 static int ext4_mb_release_context(struct ext4_allocation_context *ac)
 {
-	if (ac->ac_pa) {
-		if (ac->ac_pa->pa_linear) {
+	struct ext4_prealloc_space *pa = ac->ac_pa;
+	if (pa) {
+		if (pa->pa_linear) {
 			/* see comment in ext4_mb_use_group_pa() */
-			spin_lock(&ac->ac_pa->pa_lock);
-			ac->ac_pa->pa_pstart += ac->ac_b_ex.fe_len;
-			ac->ac_pa->pa_lstart += ac->ac_b_ex.fe_len;
-			ac->ac_pa->pa_free -= ac->ac_b_ex.fe_len;
-			ac->ac_pa->pa_len -= ac->ac_b_ex.fe_len;
-			spin_unlock(&ac->ac_pa->pa_lock);
+			spin_lock(&pa->pa_lock);
+			pa->pa_pstart += ac->ac_b_ex.fe_len;
+			pa->pa_lstart += ac->ac_b_ex.fe_len;
+			pa->pa_free -= ac->ac_b_ex.fe_len;
+			pa->pa_len -= ac->ac_b_ex.fe_len;
+			spin_unlock(&pa->pa_lock);
+			/*
+			 * We want to add the pa to the right bucket.
+			 * Remove it from the list and while adding
+			 * make sure the list to which we are adding
+			 * doesn't grow big.
+			 */
+			if (likely(pa->pa_free)) {
+				spin_lock(pa->pa_obj_lock);
+				list_del_rcu(&pa->pa_inode_list);
+				spin_unlock(pa->pa_obj_lock);
+				ext4_mb_add_n_trim(ac);
+			}
 		}
-		ext4_mb_put_pa(ac, ac->ac_sb, ac->ac_pa);
+		ext4_mb_put_pa(ac, ac->ac_sb, pa);
 	}
 	if (ac->ac_bitmap_page)
 		page_cache_release(ac->ac_bitmap_page);
@@ -4420,11 +4588,15 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
 		count -= overflow;
 	}
 	bitmap_bh = ext4_read_block_bitmap(sb, block_group);
-	if (!bitmap_bh)
+	if (!bitmap_bh) {
+		err = -EIO;
 		goto error_return;
+	}
 	gdp = ext4_get_group_desc(sb, block_group, &gd_bh);
-	if (!gdp)
+	if (!gdp) {
+		err = -EIO;
 		goto error_return;
+	}
 
 	if (in_range(ext4_block_bitmap(sb, gdp), block, count) ||
 	    in_range(ext4_inode_bitmap(sb, gdp), block, count) ||

fs/ext4/mballoc.h

@@ -164,11 +164,17 @@ struct ext4_free_extent {
  * Locality group:
  *   we try to group all related changes together
  *   so that writeback can flush/allocate them together as well
+ *   Size of lg_prealloc_list hash is determined by MB_DEFAULT_GROUP_PREALLOC
+ *   (512). We store prealloc space into the hash based on the pa_free blocks
+ *   order value.ie, fls(pa_free)-1;
  */
+#define PREALLOC_TB_SIZE 10
 struct ext4_locality_group {
 	/* for allocator */
-	struct mutex		lg_mutex;	/* to serialize allocates */
-	struct list_head	lg_prealloc_list;/* list of preallocations */
+	/* to serialize allocates */
+	struct mutex		lg_mutex;
+	/* list of preallocations */
+	struct list_head	lg_prealloc_list[PREALLOC_TB_SIZE];
 	spinlock_t		lg_prealloc_lock;
 };
 

fs/ext4/resize.c

@@ -73,7 +73,7 @@ static int verify_group_input(struct super_block *sb,
 			     "Inode bitmap not in group (block %llu)",
 			     (unsigned long long)input->inode_bitmap);
 	else if (outside(input->inode_table, start, end) ||
-	         outside(itend - 1, start, end))
+		 outside(itend - 1, start, end))
 		ext4_warning(sb, __func__,
 			     "Inode table not in group (blocks %llu-%llu)",
 			     (unsigned long long)input->inode_table, itend - 1);
@@ -104,7 +104,7 @@ static int verify_group_input(struct super_block *sb,
 			     (unsigned long long)input->inode_bitmap,
 			     start, metaend - 1);
 	else if (inside(input->inode_table, start, metaend) ||
-	         inside(itend - 1, start, metaend))
+		 inside(itend - 1, start, metaend))
 		ext4_warning(sb, __func__,
 			     "Inode table (%llu-%llu) overlaps"
 			     "GDT table (%llu-%llu)",
@@ -158,9 +158,9 @@ static int extend_or_restart_transaction(handle_t *handle, int thresh,
 	if (err) {
 		if ((err = ext4_journal_restart(handle, EXT4_MAX_TRANS_DATA)))
 			return err;
-	        if ((err = ext4_journal_get_write_access(handle, bh)))
+		if ((err = ext4_journal_get_write_access(handle, bh)))
 			return err;
-        }
+	}
 
 	return 0;
 }
@@ -416,11 +416,11 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 		       "EXT4-fs: ext4_add_new_gdb: adding group block %lu\n",
 		       gdb_num);
 
-	/*
-	 * If we are not using the primary superblock/GDT copy don't resize,
-	 * because the user tools have no way of handling this.  Probably a
-	 * bad time to do it anyways.
-	 */
+        /*
+         * If we are not using the primary superblock/GDT copy don't resize,
+         * because the user tools have no way of handling this.  Probably a
+         * bad time to do it anyways.
+         */
 	if (EXT4_SB(sb)->s_sbh->b_blocknr !=
 	    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) {
 		ext4_warning(sb, __func__,
@@ -507,14 +507,14 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	return 0;
 
 exit_inode:
-	//ext4_journal_release_buffer(handle, iloc.bh);
+	/* ext4_journal_release_buffer(handle, iloc.bh); */
 	brelse(iloc.bh);
 exit_dindj:
-	//ext4_journal_release_buffer(handle, dind);
+	/* ext4_journal_release_buffer(handle, dind); */
 exit_primary:
-	//ext4_journal_release_buffer(handle, *primary);
+	/* ext4_journal_release_buffer(handle, *primary); */
 exit_sbh:
-	//ext4_journal_release_buffer(handle, *primary);
+	/* ext4_journal_release_buffer(handle, *primary); */
 exit_dind:
 	brelse(dind);
 exit_bh:
@@ -818,12 +818,12 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 	if ((err = ext4_journal_get_write_access(handle, sbi->s_sbh)))
 		goto exit_journal;
 
-	/*
-	 * We will only either add reserved group blocks to a backup group
-	 * or remove reserved blocks for the first group in a new group block.
-	 * Doing both would be mean more complex code, and sane people don't
-	 * use non-sparse filesystems anymore.  This is already checked above.
-	 */
+        /*
+         * We will only either add reserved group blocks to a backup group
+         * or remove reserved blocks for the first group in a new group block.
+         * Doing both would be mean more complex code, and sane people don't
+         * use non-sparse filesystems anymore.  This is already checked above.
+         */
 	if (gdb_off) {
 		primary = sbi->s_group_desc[gdb_num];
 		if ((err = ext4_journal_get_write_access(handle, primary)))
@@ -835,24 +835,24 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 	} else if ((err = add_new_gdb(handle, inode, input, &primary)))
 		goto exit_journal;
 
-	/*
-	 * OK, now we've set up the new group.  Time to make it active.
-	 *
-	 * Current kernels don't lock all allocations via lock_super(),
-	 * so we have to be safe wrt. concurrent accesses the group
-	 * data.  So we need to be careful to set all of the relevant
-	 * group descriptor data etc. *before* we enable the group.
-	 *
-	 * The key field here is sbi->s_groups_count: as long as
-	 * that retains its old value, nobody is going to access the new
-	 * group.
-	 *
-	 * So first we update all the descriptor metadata for the new
-	 * group; then we update the total disk blocks count; then we
-	 * update the groups count to enable the group; then finally we
-	 * update the free space counts so that the system can start
-	 * using the new disk blocks.
-	 */
+        /*
+         * OK, now we've set up the new group.  Time to make it active.
+         *
+         * Current kernels don't lock all allocations via lock_super(),
+         * so we have to be safe wrt. concurrent accesses the group
+         * data.  So we need to be careful to set all of the relevant
+         * group descriptor data etc. *before* we enable the group.
+         *
+         * The key field here is sbi->s_groups_count: as long as
+         * that retains its old value, nobody is going to access the new
+         * group.
+         *
+         * So first we update all the descriptor metadata for the new
+         * group; then we update the total disk blocks count; then we
+         * update the groups count to enable the group; then finally we
+         * update the free space counts so that the system can start
+         * using the new disk blocks.
+         */
 
 	/* Update group descriptor block for new group */
 	gdp = (struct ext4_group_desc *)((char *)primary->b_data +
@@ -946,7 +946,8 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 	return err;
 } /* ext4_group_add */
 
-/* Extend the filesystem to the new number of blocks specified.  This entry
+/*
+ * Extend the filesystem to the new number of blocks specified.  This entry
  * point is only used to extend the current filesystem to the end of the last
  * existing group.  It can be accessed via ioctl, or by "remount,resize=<size>"
  * for emergencies (because it has no dependencies on reserved blocks).
@@ -1024,7 +1025,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 			     o_blocks_count + add, add);
 
 	/* See if the device is actually as big as what was requested */
-	bh = sb_bread(sb, o_blocks_count + add -1);
+	bh = sb_bread(sb, o_blocks_count + add - 1);
 	if (!bh) {
 		ext4_warning(sb, __func__,
 			     "can't read last block, resize aborted");

fs/ext4/super.c

@@ -49,20 +49,19 @@ static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
 			     unsigned long journal_devnum);
 static int ext4_create_journal(struct super_block *, struct ext4_super_block *,
 			       unsigned int);
-static void ext4_commit_super (struct super_block * sb,
-			       struct ext4_super_block * es,
-			       int sync);
-static void ext4_mark_recovery_complete(struct super_block * sb,
-					struct ext4_super_block * es);
-static void ext4_clear_journal_err(struct super_block * sb,
-				   struct ext4_super_block * es);
+static void ext4_commit_super(struct super_block *sb,
+			      struct ext4_super_block *es, int sync);
+static void ext4_mark_recovery_complete(struct super_block *sb,
+					struct ext4_super_block *es);
+static void ext4_clear_journal_err(struct super_block *sb,
+				   struct ext4_super_block *es);
 static int ext4_sync_fs(struct super_block *sb, int wait);
-static const char *ext4_decode_error(struct super_block * sb, int errno,
+static const char *ext4_decode_error(struct super_block *sb, int errno,
 				     char nbuf[16]);
-static int ext4_remount (struct super_block * sb, int * flags, char * data);
-static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf);
+static int ext4_remount(struct super_block *sb, int *flags, char *data);
+static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf);
 static void ext4_unlockfs(struct super_block *sb);
-static void ext4_write_super (struct super_block * sb);
+static void ext4_write_super(struct super_block *sb);
 static void ext4_write_super_lockfs(struct super_block *sb);
 
 
@@ -211,15 +210,15 @@ static void ext4_handle_error(struct super_block *sb)
 	if (sb->s_flags & MS_RDONLY)
 		return;
 
-	if (!test_opt (sb, ERRORS_CONT)) {
+	if (!test_opt(sb, ERRORS_CONT)) {
 		journal_t *journal = EXT4_SB(sb)->s_journal;
 
 		EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT;
 		if (journal)
 			jbd2_journal_abort(journal, -EIO);
 	}
-	if (test_opt (sb, ERRORS_RO)) {
-		printk (KERN_CRIT "Remounting filesystem read-only\n");
+	if (test_opt(sb, ERRORS_RO)) {
+		printk(KERN_CRIT "Remounting filesystem read-only\n");
 		sb->s_flags |= MS_RDONLY;
 	}
 	ext4_commit_super(sb, es, 1);
@@ -228,13 +227,13 @@ static void ext4_handle_error(struct super_block *sb)
 			sb->s_id);
 }
 
-void ext4_error (struct super_block * sb, const char * function,
-		 const char * fmt, ...)
+void ext4_error(struct super_block *sb, const char *function,
+		const char *fmt, ...)
 {
 	va_list args;
 
 	va_start(args, fmt);
-	printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function);
+	printk(KERN_CRIT "EXT4-fs error (device %s): %s: ", sb->s_id, function);
 	vprintk(fmt, args);
 	printk("\n");
 	va_end(args);
@@ -242,7 +241,7 @@ void ext4_error (struct super_block * sb, const char * function,
 	ext4_handle_error(sb);
 }
 
-static const char *ext4_decode_error(struct super_block * sb, int errno,
+static const char *ext4_decode_error(struct super_block *sb, int errno,
 				     char nbuf[16])
 {
 	char *errstr = NULL;
@@ -278,8 +277,7 @@ static const char *ext4_decode_error(struct super_block * sb, int errno,
 /* __ext4_std_error decodes expected errors from journaling functions
  * automatically and invokes the appropriate error response.  */
 
-void __ext4_std_error (struct super_block * sb, const char * function,
-		       int errno)
+void __ext4_std_error(struct super_block *sb, const char *function, int errno)
 {
 	char nbuf[16];
 	const char *errstr;
@@ -292,8 +290,8 @@ void __ext4_std_error (struct super_block * sb, const char * function,
 		return;
 
 	errstr = ext4_decode_error(sb, errno, nbuf);
-	printk (KERN_CRIT "EXT4-fs error (device %s) in %s: %s\n",
-		sb->s_id, function, errstr);
+	printk(KERN_CRIT "EXT4-fs error (device %s) in %s: %s\n",
+	       sb->s_id, function, errstr);
 
 	ext4_handle_error(sb);
 }
@@ -308,15 +306,15 @@ void __ext4_std_error (struct super_block * sb, const char * function,
  * case we take the easy way out and panic immediately.
  */
 
-void ext4_abort (struct super_block * sb, const char * function,
-		 const char * fmt, ...)
+void ext4_abort(struct super_block *sb, const char *function,
+		const char *fmt, ...)
 {
 	va_list args;
 
-	printk (KERN_CRIT "ext4_abort called.\n");
+	printk(KERN_CRIT "ext4_abort called.\n");
 
 	va_start(args, fmt);
-	printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function);
+	printk(KERN_CRIT "EXT4-fs error (device %s): %s: ", sb->s_id, function);
 	vprintk(fmt, args);
 	printk("\n");
 	va_end(args);
@@ -334,8 +332,8 @@ void ext4_abort (struct super_block * sb, const char * function,
 	jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO);
 }
 
-void ext4_warning (struct super_block * sb, const char * function,
-		   const char * fmt, ...)
+void ext4_warning(struct super_block *sb, const char *function,
+		  const char *fmt, ...)
 {
 	va_list args;
 
@@ -496,7 +494,7 @@ static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi)
 	}
 }
 
-static void ext4_put_super (struct super_block * sb)
+static void ext4_put_super(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_super_block *es = sbi->s_es;
@@ -647,7 +645,8 @@ static void ext4_clear_inode(struct inode *inode)
 				       &EXT4_I(inode)->jinode);
 }
 
-static inline void ext4_show_quota_options(struct seq_file *seq, struct super_block *sb)
+static inline void ext4_show_quota_options(struct seq_file *seq,
+					   struct super_block *sb)
 {
 #if defined(CONFIG_QUOTA)
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -822,8 +821,8 @@ static struct dentry *ext4_fh_to_parent(struct super_block *sb, struct fid *fid,
 }
 
 #ifdef CONFIG_QUOTA
-#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group")
-#define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
+#define QTYPE2NAME(t) ((t) == USRQUOTA?"user":"group")
+#define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
 
 static int ext4_dquot_initialize(struct inode *inode, int type);
 static int ext4_dquot_drop(struct inode *inode);
@@ -991,12 +990,12 @@ static ext4_fsblk_t get_sb_block(void **data)
 	return sb_block;
 }
 
-static int parse_options (char *options, struct super_block *sb,
-			  unsigned int *inum, unsigned long *journal_devnum,
-			  ext4_fsblk_t *n_blocks_count, int is_remount)
+static int parse_options(char *options, struct super_block *sb,
+			 unsigned int *inum, unsigned long *journal_devnum,
+			 ext4_fsblk_t *n_blocks_count, int is_remount)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	char * p;
+	char *p;
 	substring_t args[MAX_OPT_ARGS];
 	int data_opt = 0;
 	int option;
@@ -1009,7 +1008,7 @@ static int parse_options (char *options, struct super_block *sb,
 	if (!options)
 		return 1;
 
-	while ((p = strsep (&options, ",")) != NULL) {
+	while ((p = strsep(&options, ",")) != NULL) {
 		int token;
 		if (!*p)
 			continue;
@@ -1017,16 +1016,16 @@ static int parse_options (char *options, struct super_block *sb,
 		token = match_token(p, tokens, args);
 		switch (token) {
 		case Opt_bsd_df:
-			clear_opt (sbi->s_mount_opt, MINIX_DF);
+			clear_opt(sbi->s_mount_opt, MINIX_DF);
 			break;
 		case Opt_minix_df:
-			set_opt (sbi->s_mount_opt, MINIX_DF);
+			set_opt(sbi->s_mount_opt, MINIX_DF);
 			break;
 		case Opt_grpid:
-			set_opt (sbi->s_mount_opt, GRPID);
+			set_opt(sbi->s_mount_opt, GRPID);
 			break;
 		case Opt_nogrpid:
-			clear_opt (sbi->s_mount_opt, GRPID);
+			clear_opt(sbi->s_mount_opt, GRPID);
 			break;
 		case Opt_resuid:
 			if (match_int(&args[0], &option))
@@ -1043,41 +1042,41 @@ static int parse_options (char *options, struct super_block *sb,
 			/* *sb_block = match_int(&args[0]); */
 			break;
 		case Opt_err_panic:
-			clear_opt (sbi->s_mount_opt, ERRORS_CONT);
-			clear_opt (sbi->s_mount_opt, ERRORS_RO);
-			set_opt (sbi->s_mount_opt, ERRORS_PANIC);
+			clear_opt(sbi->s_mount_opt, ERRORS_CONT);
+			clear_opt(sbi->s_mount_opt, ERRORS_RO);
+			set_opt(sbi->s_mount_opt, ERRORS_PANIC);
 			break;
 		case Opt_err_ro:
-			clear_opt (sbi->s_mount_opt, ERRORS_CONT);
-			clear_opt (sbi->s_mount_opt, ERRORS_PANIC);
-			set_opt (sbi->s_mount_opt, ERRORS_RO);
+			clear_opt(sbi->s_mount_opt, ERRORS_CONT);
+			clear_opt(sbi->s_mount_opt, ERRORS_PANIC);
+			set_opt(sbi->s_mount_opt, ERRORS_RO);
 			break;
 		case Opt_err_cont:
-			clear_opt (sbi->s_mount_opt, ERRORS_RO);
-			clear_opt (sbi->s_mount_opt, ERRORS_PANIC);
-			set_opt (sbi->s_mount_opt, ERRORS_CONT);
+			clear_opt(sbi->s_mount_opt, ERRORS_RO);
+			clear_opt(sbi->s_mount_opt, ERRORS_PANIC);
+			set_opt(sbi->s_mount_opt, ERRORS_CONT);
 			break;
 		case Opt_nouid32:
-			set_opt (sbi->s_mount_opt, NO_UID32);
+			set_opt(sbi->s_mount_opt, NO_UID32);
 			break;
 		case Opt_nocheck:
-			clear_opt (sbi->s_mount_opt, CHECK);
+			clear_opt(sbi->s_mount_opt, CHECK);
 			break;
 		case Opt_debug:
-			set_opt (sbi->s_mount_opt, DEBUG);
+			set_opt(sbi->s_mount_opt, DEBUG);
 			break;
 		case Opt_oldalloc:
-			set_opt (sbi->s_mount_opt, OLDALLOC);
+			set_opt(sbi->s_mount_opt, OLDALLOC);
 			break;
 		case Opt_orlov:
-			clear_opt (sbi->s_mount_opt, OLDALLOC);
+			clear_opt(sbi->s_mount_opt, OLDALLOC);
 			break;
 #ifdef CONFIG_EXT4DEV_FS_XATTR
 		case Opt_user_xattr:
-			set_opt (sbi->s_mount_opt, XATTR_USER);
+			set_opt(sbi->s_mount_opt, XATTR_USER);
 			break;
 		case Opt_nouser_xattr:
-			clear_opt (sbi->s_mount_opt, XATTR_USER);
+			clear_opt(sbi->s_mount_opt, XATTR_USER);
 			break;
 #else
 		case Opt_user_xattr:
@@ -1115,7 +1114,7 @@ static int parse_options (char *options, struct super_block *sb,
 				       "journal on remount\n");
 				return 0;
 			}
-			set_opt (sbi->s_mount_opt, UPDATE_JOURNAL);
+			set_opt(sbi->s_mount_opt, UPDATE_JOURNAL);
 			break;
 		case Opt_journal_inum:
 			if (is_remount) {
@@ -1145,7 +1144,7 @@ static int parse_options (char *options, struct super_block *sb,
 			set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM);
 			break;
 		case Opt_noload:
-			set_opt (sbi->s_mount_opt, NOLOAD);
+			set_opt(sbi->s_mount_opt, NOLOAD);
 			break;
 		case Opt_commit:
 			if (match_int(&args[0], &option))
@@ -1331,7 +1330,7 @@ static int parse_options (char *options, struct super_block *sb,
 					"on this filesystem, use tune2fs\n");
 				return 0;
 			}
-			set_opt (sbi->s_mount_opt, EXTENTS);
+			set_opt(sbi->s_mount_opt, EXTENTS);
 			break;
 		case Opt_noextents:
 			/*
@@ -1348,7 +1347,7 @@ static int parse_options (char *options, struct super_block *sb,
 						"-o noextents options\n");
 				return 0;
 			}
-			clear_opt (sbi->s_mount_opt, EXTENTS);
+			clear_opt(sbi->s_mount_opt, EXTENTS);
 			break;
 		case Opt_i_version:
 			set_opt(sbi->s_mount_opt, I_VERSION);
@@ -1374,9 +1373,9 @@ static int parse_options (char *options, struct super_block *sb,
 			set_opt(sbi->s_mount_opt, DELALLOC);
 			break;
 		default:
-			printk (KERN_ERR
-				"EXT4-fs: Unrecognized mount option \"%s\" "
-				"or missing value\n", p);
+			printk(KERN_ERR
+			       "EXT4-fs: Unrecognized mount option \"%s\" "
+			       "or missing value\n", p);
 			return 0;
 		}
 	}
@@ -1423,31 +1422,31 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
 	int res = 0;
 
 	if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) {
-		printk (KERN_ERR "EXT4-fs warning: revision level too high, "
-			"forcing read-only mode\n");
+		printk(KERN_ERR "EXT4-fs warning: revision level too high, "
+		       "forcing read-only mode\n");
 		res = MS_RDONLY;
 	}
 	if (read_only)
 		return res;
 	if (!(sbi->s_mount_state & EXT4_VALID_FS))
-		printk (KERN_WARNING "EXT4-fs warning: mounting unchecked fs, "
-			"running e2fsck is recommended\n");
+		printk(KERN_WARNING "EXT4-fs warning: mounting unchecked fs, "
+		       "running e2fsck is recommended\n");
 	else if ((sbi->s_mount_state & EXT4_ERROR_FS))
-		printk (KERN_WARNING
-			"EXT4-fs warning: mounting fs with errors, "
-			"running e2fsck is recommended\n");
+		printk(KERN_WARNING
+		       "EXT4-fs warning: mounting fs with errors, "
+		       "running e2fsck is recommended\n");
 	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 &&
 		 le16_to_cpu(es->s_mnt_count) >=
 		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
-		printk (KERN_WARNING
-			"EXT4-fs warning: maximal mount count reached, "
-			"running e2fsck is recommended\n");
+		printk(KERN_WARNING
+		       "EXT4-fs warning: maximal mount count reached, "
+		       "running e2fsck is recommended\n");
 	else if (le32_to_cpu(es->s_checkinterval) &&
 		(le32_to_cpu(es->s_lastcheck) +
 			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
-		printk (KERN_WARNING
-			"EXT4-fs warning: checktime reached, "
-			"running e2fsck is recommended\n");
+		printk(KERN_WARNING
+		       "EXT4-fs warning: checktime reached, "
+		       "running e2fsck is recommended\n");
 #if 0
 		/* @@@ We _will_ want to clear the valid bit if we find
 		 * inconsistencies, to force a fsck at reboot.  But for
@@ -1506,14 +1505,13 @@ static int ext4_fill_flex_info(struct super_block *sb)
 
 	flex_group_count = (sbi->s_groups_count + groups_per_flex - 1) /
 		groups_per_flex;
-	sbi->s_flex_groups = kmalloc(flex_group_count *
+	sbi->s_flex_groups = kzalloc(flex_group_count *
 				     sizeof(struct flex_groups), GFP_KERNEL);
 	if (sbi->s_flex_groups == NULL) {
-		printk(KERN_ERR "EXT4-fs: not enough memory\n");
+		printk(KERN_ERR "EXT4-fs: not enough memory for "
+				"%lu flex groups\n", flex_group_count);
 		goto failed;
 	}
-	memset(sbi->s_flex_groups, 0, flex_group_count *
-	       sizeof(struct flex_groups));
 
 	gdp = ext4_get_group_desc(sb, 1, &bh);
 	block_bitmap = ext4_block_bitmap(sb, gdp) - 1;
@@ -1597,16 +1595,14 @@ static int ext4_check_descriptors(struct super_block *sb)
 				(EXT4_BLOCKS_PER_GROUP(sb) - 1);
 
 		block_bitmap = ext4_block_bitmap(sb, gdp);
-		if (block_bitmap < first_block || block_bitmap > last_block)
-		{
+		if (block_bitmap < first_block || block_bitmap > last_block) {
 			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
 			       "Block bitmap for group %lu not in group "
 			       "(block %llu)!", i, block_bitmap);
 			return 0;
 		}
 		inode_bitmap = ext4_inode_bitmap(sb, gdp);
-		if (inode_bitmap < first_block || inode_bitmap > last_block)
-		{
+		if (inode_bitmap < first_block || inode_bitmap > last_block) {
 			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
 			       "Inode bitmap for group %lu not in group "
 			       "(block %llu)!", i, inode_bitmap);
@@ -1614,26 +1610,28 @@ static int ext4_check_descriptors(struct super_block *sb)
 		}
 		inode_table = ext4_inode_table(sb, gdp);
 		if (inode_table < first_block ||
-		    inode_table + sbi->s_itb_per_group - 1 > last_block)
-		{
+		    inode_table + sbi->s_itb_per_group - 1 > last_block) {
 			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
 			       "Inode table for group %lu not in group "
 			       "(block %llu)!", i, inode_table);
 			return 0;
 		}
+		spin_lock(sb_bgl_lock(sbi, i));
 		if (!ext4_group_desc_csum_verify(sbi, i, gdp)) {
 			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
 			       "Checksum for group %lu failed (%u!=%u)\n",
 			       i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
 			       gdp)), le16_to_cpu(gdp->bg_checksum));
-			return 0;
+			if (!(sb->s_flags & MS_RDONLY))
+				return 0;
 		}
+		spin_unlock(sb_bgl_lock(sbi, i));
 		if (!flexbg_flag)
 			first_block += EXT4_BLOCKS_PER_GROUP(sb);
 	}
 
 	ext4_free_blocks_count_set(sbi->s_es, ext4_count_free_blocks(sb));
-	sbi->s_es->s_free_inodes_count=cpu_to_le32(ext4_count_free_inodes(sb));
+	sbi->s_es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb));
 	return 1;
 }
 
@@ -1654,8 +1652,8 @@ static int ext4_check_descriptors(struct super_block *sb)
  * e2fsck was run on this filesystem, and it must have already done the orphan
  * inode cleanup for us, so we can safely abort without any further action.
  */
-static void ext4_orphan_cleanup (struct super_block * sb,
-				 struct ext4_super_block * es)
+static void ext4_orphan_cleanup(struct super_block *sb,
+				struct ext4_super_block *es)
 {
 	unsigned int s_flags = sb->s_flags;
 	int nr_orphans = 0, nr_truncates = 0;
@@ -1732,7 +1730,7 @@ static void ext4_orphan_cleanup (struct super_block * sb,
 		iput(inode);  /* The delete magic happens here! */
 	}
 
-#define PLURAL(x) (x), ((x)==1) ? "" : "s"
+#define PLURAL(x) (x), ((x) == 1) ? "" : "s"
 
 	if (nr_orphans)
 		printk(KERN_INFO "EXT4-fs: %s: %d orphan inode%s deleted\n",
@@ -1899,12 +1897,12 @@ static unsigned long ext4_get_stripe_size(struct ext4_sb_info *sbi)
 	return 0;
 }
 
-static int ext4_fill_super (struct super_block *sb, void *data, int silent)
+static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 				__releases(kernel_lock)
 				__acquires(kernel_lock)
 
 {
-	struct buffer_head * bh;
+	struct buffer_head *bh;
 	struct ext4_super_block *es = NULL;
 	struct ext4_sb_info *sbi;
 	ext4_fsblk_t block;
@@ -1953,7 +1951,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 	}
 
 	if (!(bh = sb_bread(sb, logical_sb_block))) {
-		printk (KERN_ERR "EXT4-fs: unable to read superblock\n");
+		printk(KERN_ERR "EXT4-fs: unable to read superblock\n");
 		goto out_fail;
 	}
 	/*
@@ -2026,8 +2024,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 	set_opt(sbi->s_mount_opt, DELALLOC);
 
 
-	if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum,
-			    NULL, 0))
+	if (!parse_options((char *) data, sb, &journal_inum, &journal_devnum,
+			   NULL, 0))
 		goto failed_mount;
 
 	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
@@ -2102,7 +2100,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 			goto failed_mount;
 		}
 
-		brelse (bh);
+		brelse(bh);
 		logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE;
 		offset = do_div(logical_sb_block, blocksize);
 		bh = sb_bread(sb, logical_sb_block);
@@ -2114,8 +2112,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		es = (struct ext4_super_block *)(((char *)bh->b_data) + offset);
 		sbi->s_es = es;
 		if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) {
-			printk (KERN_ERR
-				"EXT4-fs: Magic mismatch, very weird !\n");
+			printk(KERN_ERR
+			       "EXT4-fs: Magic mismatch, very weird !\n");
 			goto failed_mount;
 		}
 	}
@@ -2132,9 +2130,9 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) ||
 		    (!is_power_of_2(sbi->s_inode_size)) ||
 		    (sbi->s_inode_size > blocksize)) {
-			printk (KERN_ERR
-				"EXT4-fs: unsupported inode size: %d\n",
-				sbi->s_inode_size);
+			printk(KERN_ERR
+			       "EXT4-fs: unsupported inode size: %d\n",
+			       sbi->s_inode_size);
 			goto failed_mount;
 		}
 		if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE)
@@ -2166,20 +2164,20 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 	sbi->s_mount_state = le16_to_cpu(es->s_state);
 	sbi->s_addr_per_block_bits = ilog2(EXT4_ADDR_PER_BLOCK(sb));
 	sbi->s_desc_per_block_bits = ilog2(EXT4_DESC_PER_BLOCK(sb));
-	for (i=0; i < 4; i++)
+	for (i = 0; i < 4; i++)
 		sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]);
 	sbi->s_def_hash_version = es->s_def_hash_version;
 
 	if (sbi->s_blocks_per_group > blocksize * 8) {
-		printk (KERN_ERR
-			"EXT4-fs: #blocks per group too big: %lu\n",
-			sbi->s_blocks_per_group);
+		printk(KERN_ERR
+		       "EXT4-fs: #blocks per group too big: %lu\n",
+		       sbi->s_blocks_per_group);
 		goto failed_mount;
 	}
 	if (sbi->s_inodes_per_group > blocksize * 8) {
-		printk (KERN_ERR
-			"EXT4-fs: #inodes per group too big: %lu\n",
-			sbi->s_inodes_per_group);
+		printk(KERN_ERR
+		       "EXT4-fs: #inodes per group too big: %lu\n",
+		       sbi->s_inodes_per_group);
 		goto failed_mount;
 	}
 
@@ -2213,10 +2211,10 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 	sbi->s_groups_count = blocks_count;
 	db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
 		   EXT4_DESC_PER_BLOCK(sb);
-	sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *),
+	sbi->s_group_desc = kmalloc(db_count * sizeof(struct buffer_head *),
 				    GFP_KERNEL);
 	if (sbi->s_group_desc == NULL) {
-		printk (KERN_ERR "EXT4-fs: not enough memory\n");
+		printk(KERN_ERR "EXT4-fs: not enough memory\n");
 		goto failed_mount;
 	}
 
@@ -2226,13 +2224,13 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		block = descriptor_loc(sb, logical_sb_block, i);
 		sbi->s_group_desc[i] = sb_bread(sb, block);
 		if (!sbi->s_group_desc[i]) {
-			printk (KERN_ERR "EXT4-fs: "
-				"can't read group descriptor %d\n", i);
+			printk(KERN_ERR "EXT4-fs: "
+			       "can't read group descriptor %d\n", i);
 			db_count = i;
 			goto failed_mount2;
 		}
 	}
-	if (!ext4_check_descriptors (sb)) {
+	if (!ext4_check_descriptors(sb)) {
 		printk(KERN_ERR "EXT4-fs: group descriptors corrupted!\n");
 		goto failed_mount2;
 	}
@@ -2308,11 +2306,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		    EXT4_SB(sb)->s_journal->j_failed_commit) {
 			printk(KERN_CRIT "EXT4-fs error (device %s): "
 			       "ext4_fill_super: Journal transaction "
-			       "%u is corrupt\n", sb->s_id, 
+			       "%u is corrupt\n", sb->s_id,
 			       EXT4_SB(sb)->s_journal->j_failed_commit);
-			if (test_opt (sb, ERRORS_RO)) {
-				printk (KERN_CRIT
-					"Mounting filesystem read-only\n");
+			if (test_opt(sb, ERRORS_RO)) {
+				printk(KERN_CRIT
+				       "Mounting filesystem read-only\n");
 				sb->s_flags |= MS_RDONLY;
 				EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
 				es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
@@ -2332,9 +2330,9 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 			goto failed_mount3;
 	} else {
 		if (!silent)
-			printk (KERN_ERR
-				"ext4: No journal on filesystem on %s\n",
-				sb->s_id);
+			printk(KERN_ERR
+			       "ext4: No journal on filesystem on %s\n",
+			       sb->s_id);
 		goto failed_mount3;
 	}
 
@@ -2418,7 +2416,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 		goto failed_mount4;
 	}
 
-	ext4_setup_super (sb, es, sb->s_flags & MS_RDONLY);
+	ext4_setup_super(sb, es, sb->s_flags & MS_RDONLY);
 
 	/* determine the minimum size of new large inodes, if present */
 	if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) {
@@ -2457,12 +2455,12 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent)
 	ext4_orphan_cleanup(sb, es);
 	EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS;
 	if (needs_recovery)
-		printk (KERN_INFO "EXT4-fs: recovery complete.\n");
+		printk(KERN_INFO "EXT4-fs: recovery complete.\n");
 	ext4_mark_recovery_complete(sb, es);
-	printk (KERN_INFO "EXT4-fs: mounted filesystem with %s data mode.\n",
-		test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ? "journal":
-		test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered":
-		"writeback");
+	printk(KERN_INFO "EXT4-fs: mounted filesystem with %s data mode.\n",
+	       test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ? "journal":
+	       test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered":
+	       "writeback");
 
 	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) {
 		printk(KERN_WARNING "EXT4-fs: Ignoring delalloc option - "
@@ -2575,14 +2573,14 @@ static journal_t *ext4_get_journal(struct super_block *sb,
 static journal_t *ext4_get_dev_journal(struct super_block *sb,
 				       dev_t j_dev)
 {
-	struct buffer_head * bh;
+	struct buffer_head *bh;
 	journal_t *journal;
 	ext4_fsblk_t start;
 	ext4_fsblk_t len;
 	int hblock, blocksize;
 	ext4_fsblk_t sb_block;
 	unsigned long offset;
-	struct ext4_super_block * es;
+	struct ext4_super_block *es;
 	struct block_device *bdev;
 
 	bdev = ext4_blkdev_get(j_dev);
@@ -2697,8 +2695,8 @@ static int ext4_load_journal(struct super_block *sb,
 					"unavailable, cannot proceed.\n");
 				return -EROFS;
 			}
-			printk (KERN_INFO "EXT4-fs: write access will "
-					"be enabled during recovery.\n");
+			printk(KERN_INFO "EXT4-fs: write access will "
+			       "be enabled during recovery.\n");
 		}
 	}
 
@@ -2751,8 +2749,8 @@ static int ext4_load_journal(struct super_block *sb,
 	return 0;
 }
 
-static int ext4_create_journal(struct super_block * sb,
-			       struct ext4_super_block * es,
+static int ext4_create_journal(struct super_block *sb,
+			       struct ext4_super_block *es,
 			       unsigned int journal_inum)
 {
 	journal_t *journal;
@@ -2793,9 +2791,8 @@ static int ext4_create_journal(struct super_block * sb,
 	return 0;
 }
 
-static void ext4_commit_super (struct super_block * sb,
-			       struct ext4_super_block * es,
-			       int sync)
+static void ext4_commit_super(struct super_block *sb,
+			      struct ext4_super_block *es, int sync)
 {
 	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
 
@@ -2816,8 +2813,8 @@ static void ext4_commit_super (struct super_block * sb,
  * remounting) the filesystem readonly, then we will end up with a
  * consistent fs on disk.  Record that fact.
  */
-static void ext4_mark_recovery_complete(struct super_block * sb,
-					struct ext4_super_block * es)
+static void ext4_mark_recovery_complete(struct super_block *sb,
+					struct ext4_super_block *es)
 {
 	journal_t *journal = EXT4_SB(sb)->s_journal;
 
@@ -2839,8 +2836,8 @@ static void ext4_mark_recovery_complete(struct super_block * sb,
  * has recorded an error from a previous lifetime, move that error to the
  * main filesystem now.
  */
-static void ext4_clear_journal_err(struct super_block * sb,
-				   struct ext4_super_block * es)
+static void ext4_clear_journal_err(struct super_block *sb,
+				   struct ext4_super_block *es)
 {
 	journal_t *journal;
 	int j_errno;
@@ -2865,7 +2862,7 @@ static void ext4_clear_journal_err(struct super_block * sb,
 
 		EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
 		es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
-		ext4_commit_super (sb, es, 1);
+		ext4_commit_super(sb, es, 1);
 
 		jbd2_journal_clear_err(journal);
 	}
@@ -2898,7 +2895,7 @@ int ext4_force_commit(struct super_block *sb)
  * This implicitly triggers the writebehind on sync().
  */
 
-static void ext4_write_super (struct super_block * sb)
+static void ext4_write_super(struct super_block *sb)
 {
 	if (mutex_trylock(&sb->s_lock) != 0)
 		BUG();
@@ -2954,13 +2951,14 @@ static void ext4_unlockfs(struct super_block *sb)
 	}
 }
 
-static int ext4_remount (struct super_block * sb, int * flags, char * data)
+static int ext4_remount(struct super_block *sb, int *flags, char *data)
 {
-	struct ext4_super_block * es;
+	struct ext4_super_block *es;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	ext4_fsblk_t n_blocks_count = 0;
 	unsigned long old_sb_flags;
 	struct ext4_mount_options old_opts;
+	ext4_group_t g;
 	int err;
 #ifdef CONFIG_QUOTA
 	int i;
@@ -3038,6 +3036,26 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data)
 				goto restore_opts;
 			}
 
+			/*
+			 * Make sure the group descriptor checksums
+			 * are sane.  If they aren't, refuse to
+			 * remount r/w.
+			 */
+			for (g = 0; g < sbi->s_groups_count; g++) {
+				struct ext4_group_desc *gdp =
+					ext4_get_group_desc(sb, g, NULL);
+
+				if (!ext4_group_desc_csum_verify(sbi, g, gdp)) {
+					printk(KERN_ERR
+	       "EXT4-fs: ext4_remount: "
+		"Checksum for group %lu failed (%u!=%u)\n",
+		g, le16_to_cpu(ext4_group_desc_csum(sbi, g, gdp)),
+					       le16_to_cpu(gdp->bg_checksum));
+					err = -EINVAL;
+					goto restore_opts;
+				}
+			}
+
 			/*
 			 * If we have an unprocessed orphan list hanging
 			 * around from a previously readonly bdev mount,
@@ -3063,7 +3081,7 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data)
 			sbi->s_mount_state = le16_to_cpu(es->s_state);
 			if ((err = ext4_group_extend(sb, es, n_blocks_count)))
 				goto restore_opts;
-			if (!ext4_setup_super (sb, es, 0))
+			if (!ext4_setup_super(sb, es, 0))
 				sb->s_flags &= ~MS_RDONLY;
 		}
 	}
@@ -3093,7 +3111,7 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data)
 	return err;
 }
 
-static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf)
+static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *sb = dentry->d_sb;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -3331,12 +3349,12 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id,
 	}
 	/* Journaling quota? */
 	if (EXT4_SB(sb)->s_qf_names[type]) {
-		/* Quotafile not of fs root? */
+		/* Quotafile not in fs root? */
 		if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode)
 			printk(KERN_WARNING
 				"EXT4-fs: Quota file not on filesystem root. "
 				"Journaled quota will not work.\n");
- 	}
+	}
 
 	/*
 	 * When we journal data on quota file, we have to flush journal to see

fs/ext4/xattr.c

@@ -1512,7 +1512,7 @@ static inline void ext4_xattr_hash_entry(struct ext4_xattr_header *header,
 	char *name = entry->e_name;
 	int n;
 
-	for (n=0; n < entry->e_name_len; n++) {
+	for (n = 0; n < entry->e_name_len; n++) {
 		hash = (hash << NAME_HASH_SHIFT) ^
 		       (hash >> (8*sizeof(hash) - NAME_HASH_SHIFT)) ^
 		       *name++;

fs/jbd2/commit.c

@@ -262,8 +262,18 @@ static int journal_finish_inode_data_buffers(journal_t *journal,
 		jinode->i_flags |= JI_COMMIT_RUNNING;
 		spin_unlock(&journal->j_list_lock);
 		err = filemap_fdatawait(jinode->i_vfs_inode->i_mapping);
-		if (!ret)
-			ret = err;
+		if (err) {
+			/*
+			 * Because AS_EIO is cleared by
+			 * wait_on_page_writeback_range(), set it again so
+			 * that user process can get -EIO from fsync().
+			 */
+			set_bit(AS_EIO,
+				&jinode->i_vfs_inode->i_mapping->flags);
+
+			if (!ret)
+				ret = err;
+		}
 		spin_lock(&journal->j_list_lock);
 		jinode->i_flags &= ~JI_COMMIT_RUNNING;
 		wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING);
@@ -670,8 +680,14 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 * commit block, which happens below in such setting.
 	 */
 	err = journal_finish_inode_data_buffers(journal, commit_transaction);
-	if (err)
-		jbd2_journal_abort(journal, err);
+	if (err) {
+		char b[BDEVNAME_SIZE];
+
+		printk(KERN_WARNING
+			"JBD2: Detected IO errors while flushing file data "
+			"on %s\n", bdevname(journal->j_fs_dev, b));
+		err = 0;
+	}
 
 	/* Lo and behold: we have just managed to send a transaction to
            the log.  Before we can commit it, wait for the IO so far to

fs/jbd2/journal.c

@@ -68,7 +68,6 @@ EXPORT_SYMBOL(jbd2_journal_set_features);
 EXPORT_SYMBOL(jbd2_journal_create);
 EXPORT_SYMBOL(jbd2_journal_load);
 EXPORT_SYMBOL(jbd2_journal_destroy);
-EXPORT_SYMBOL(jbd2_journal_update_superblock);
 EXPORT_SYMBOL(jbd2_journal_abort);
 EXPORT_SYMBOL(jbd2_journal_errno);
 EXPORT_SYMBOL(jbd2_journal_ack_err);