Merge tag 'for-f2fs-3.17-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs bug fixes from Jaegeuk Kim:
 "This series includes patches to:

   - fix recovery routines
   - fix bugs related to inline_data/xattr
   - fix when casting the dentry names
   - handle EIO or ENOMEM correctly
   - fix memory leak
   - fix lock coverage"

* tag 'for-f2fs-3.17-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (28 commits)
  f2fs: reposition unlock_new_inode to prevent accessing invalid inode
  f2fs: fix wrong casting for dentry name
  f2fs: simplify by using a literal
  f2fs: truncate stale block for inline_data
  f2fs: use macro for code readability
  f2fs: introduce need_do_checkpoint for readability
  f2fs: fix incorrect calculation with total/free inode num
  f2fs: remove rename and use rename2
  f2fs: skip if inline_data was converted already
  f2fs: remove rewrite_node_page
  f2fs: avoid double lock in truncate_blocks
  f2fs: prevent checkpoint during roll-forward
  f2fs: add WARN_ON in f2fs_bug_on
  f2fs: handle EIO not to break fs consistency
  f2fs: check s_dirty under cp_mutex
  f2fs: unlock_page when node page is redirtied out
  f2fs: introduce f2fs_cp_error for readability
  f2fs: give a chance to mount again when encountering errors
  f2fs: trigger release_dirty_inode in f2fs_put_super
  f2fs: don't skip checkpoint if there is no dirty node pages
  ...

fs/f2fs/Kconfig

@@ -23,7 +23,7 @@ config F2FS_STAT_FS
 	  mounted as f2fs. Each file shows the whole f2fs information.
 
 	  /sys/kernel/debug/f2fs/status includes:
-	    - major file system information managed by f2fs currently
+	    - major filesystem information managed by f2fs currently
 	    - average SIT information about whole segments
 	    - current memory footprint consumed by f2fs.
 
@@ -68,6 +68,6 @@ config F2FS_CHECK_FS
 	bool "F2FS consistency checking feature"
 	depends on F2FS_FS
 	help
-	  Enables BUG_ONs which check the file system consistency in runtime.
+	  Enables BUG_ONs which check the filesystem consistency in runtime.
 
 	  If you want to improve the performance, say N.

fs/f2fs/checkpoint.c

@@ -160,14 +160,11 @@ static int f2fs_write_meta_page(struct page *page,
 		goto redirty_out;
 	if (wbc->for_reclaim)
 		goto redirty_out;
-
+	if (unlikely(f2fs_cp_error(sbi)))
-	/* Should not write any meta pages, if any IO error was occurred */
+		goto redirty_out;
-	if (unlikely(is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ERROR_FLAG)))
-		goto no_write;
 
 	f2fs_wait_on_page_writeback(page, META);
 	write_meta_page(sbi, page);
-no_write:
 	dec_page_count(sbi, F2FS_DIRTY_META);
 	unlock_page(page);
 	return 0;
@@ -348,7 +345,7 @@ bool exist_written_data(struct f2fs_sb_info *sbi, nid_t ino, int mode)
 	return e ? true : false;
 }
 
-static void release_dirty_inode(struct f2fs_sb_info *sbi)
+void release_dirty_inode(struct f2fs_sb_info *sbi)
 {
 	struct ino_entry *e, *tmp;
 	int i;
@@ -446,8 +443,8 @@ static void write_orphan_inodes(struct f2fs_sb_info *sbi, block_t start_blk)
 	struct f2fs_orphan_block *orphan_blk = NULL;
 	unsigned int nentries = 0;
 	unsigned short index;
-	unsigned short orphan_blocks = (unsigned short)((sbi->n_orphans +
+	unsigned short orphan_blocks =
-		(F2FS_ORPHANS_PER_BLOCK - 1)) / F2FS_ORPHANS_PER_BLOCK);
+			(unsigned short)GET_ORPHAN_BLOCKS(sbi->n_orphans);
 	struct page *page = NULL;
 	struct ino_entry *orphan = NULL;
 
@@ -737,7 +734,7 @@ retry:
 /*
  * Freeze all the FS-operations for checkpoint.
  */
-static void block_operations(struct f2fs_sb_info *sbi)
+static int block_operations(struct f2fs_sb_info *sbi)
 {
 	struct writeback_control wbc = {
 		.sync_mode = WB_SYNC_ALL,
@@ -745,6 +742,7 @@ static void block_operations(struct f2fs_sb_info *sbi)
 		.for_reclaim = 0,
 	};
 	struct blk_plug plug;
+	int err = 0;
 
 	blk_start_plug(&plug);
 
@@ -754,11 +752,15 @@ retry_flush_dents:
 	if (get_pages(sbi, F2FS_DIRTY_DENTS)) {
 		f2fs_unlock_all(sbi);
 		sync_dirty_dir_inodes(sbi);
+		if (unlikely(f2fs_cp_error(sbi))) {
+			err = -EIO;
+			goto out;
+		}
 		goto retry_flush_dents;
 	}
 
 	/*
-	 * POR: we should ensure that there is no dirty node pages
+	 * POR: we should ensure that there are no dirty node pages
 	 * until finishing nat/sit flush.
 	 */
 retry_flush_nodes:
@@ -767,9 +769,16 @@ retry_flush_nodes:
 	if (get_pages(sbi, F2FS_DIRTY_NODES)) {
 		up_write(&sbi->node_write);
 		sync_node_pages(sbi, 0, &wbc);
+		if (unlikely(f2fs_cp_error(sbi))) {
+			f2fs_unlock_all(sbi);
+			err = -EIO;
+			goto out;
+		}
 		goto retry_flush_nodes;
 	}
+out:
 	blk_finish_plug(&plug);
+	return err;
 }
 
 static void unblock_operations(struct f2fs_sb_info *sbi)
@@ -813,8 +822,11 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	discard_next_dnode(sbi, NEXT_FREE_BLKADDR(sbi, curseg));
 
 	/* Flush all the NAT/SIT pages */
-	while (get_pages(sbi, F2FS_DIRTY_META))
+	while (get_pages(sbi, F2FS_DIRTY_META)) {
 		sync_meta_pages(sbi, META, LONG_MAX);
+		if (unlikely(f2fs_cp_error(sbi)))
+			return;
+	}
 
 	next_free_nid(sbi, &last_nid);
 
@@ -825,7 +837,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	ckpt->elapsed_time = cpu_to_le64(get_mtime(sbi));
 	ckpt->valid_block_count = cpu_to_le64(valid_user_blocks(sbi));
 	ckpt->free_segment_count = cpu_to_le32(free_segments(sbi));
-	for (i = 0; i < 3; i++) {
+	for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) {
 		ckpt->cur_node_segno[i] =
 			cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_NODE));
 		ckpt->cur_node_blkoff[i] =
@@ -833,7 +845,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 		ckpt->alloc_type[i + CURSEG_HOT_NODE] =
 				curseg_alloc_type(sbi, i + CURSEG_HOT_NODE);
 	}
-	for (i = 0; i < 3; i++) {
+	for (i = 0; i < NR_CURSEG_DATA_TYPE; i++) {
 		ckpt->cur_data_segno[i] =
 			cpu_to_le32(curseg_segno(sbi, i + CURSEG_HOT_DATA));
 		ckpt->cur_data_blkoff[i] =
@@ -848,24 +860,23 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 
 	/* 2 cp  + n data seg summary + orphan inode blocks */
 	data_sum_blocks = npages_for_summary_flush(sbi);
-	if (data_sum_blocks < 3)
+	if (data_sum_blocks < NR_CURSEG_DATA_TYPE)
 		set_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
 	else
 		clear_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
 
-	orphan_blocks = (sbi->n_orphans + F2FS_ORPHANS_PER_BLOCK - 1)
+	orphan_blocks = GET_ORPHAN_BLOCKS(sbi->n_orphans);
-					/ F2FS_ORPHANS_PER_BLOCK;
 	ckpt->cp_pack_start_sum = cpu_to_le32(1 + cp_payload_blks +
 			orphan_blocks);
 
 	if (is_umount) {
 		set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
-		ckpt->cp_pack_total_block_count = cpu_to_le32(2 +
+		ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS+
 				cp_payload_blks + data_sum_blocks +
 				orphan_blocks + NR_CURSEG_NODE_TYPE);
 	} else {
 		clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
-		ckpt->cp_pack_total_block_count = cpu_to_le32(2 +
+		ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS +
 				cp_payload_blks + data_sum_blocks +
 				orphan_blocks);
 	}
@@ -924,6 +935,9 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	/* wait for previous submitted node/meta pages writeback */
 	wait_on_all_pages_writeback(sbi);
 
+	if (unlikely(f2fs_cp_error(sbi)))
+		return;
+
 	filemap_fdatawait_range(NODE_MAPPING(sbi), 0, LONG_MAX);
 	filemap_fdatawait_range(META_MAPPING(sbi), 0, LONG_MAX);
 
@@ -934,15 +948,17 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	/* Here, we only have one bio having CP pack */
 	sync_meta_pages(sbi, META_FLUSH, LONG_MAX);
 
-	if (!is_set_ckpt_flags(ckpt, CP_ERROR_FLAG)) {
-		clear_prefree_segments(sbi);
 	release_dirty_inode(sbi);
+
+	if (unlikely(f2fs_cp_error(sbi)))
+		return;
+
+	clear_prefree_segments(sbi);
 	F2FS_RESET_SB_DIRT(sbi);
-	}
 }
 
 /*
- * We guarantee that this checkpoint procedure should not fail.
+ * We guarantee that this checkpoint procedure will not fail.
  */
 void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 {
@@ -952,7 +968,13 @@ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	trace_f2fs_write_checkpoint(sbi->sb, is_umount, "start block_ops");
 
 	mutex_lock(&sbi->cp_mutex);
-	block_operations(sbi);
+
+	if (!sbi->s_dirty)
+		goto out;
+	if (unlikely(f2fs_cp_error(sbi)))
+		goto out;
+	if (block_operations(sbi))
+		goto out;
 
 	trace_f2fs_write_checkpoint(sbi->sb, is_umount, "finish block_ops");
 
@@ -976,9 +998,9 @@ void write_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	do_checkpoint(sbi, is_umount);
 
 	unblock_operations(sbi);
-	mutex_unlock(&sbi->cp_mutex);
-
 	stat_inc_cp_count(sbi->stat_info);
+out:
+	mutex_unlock(&sbi->cp_mutex);
 	trace_f2fs_write_checkpoint(sbi->sb, is_umount, "finish checkpoint");
 }
 
@@ -999,8 +1021,8 @@ void init_ino_entry_info(struct f2fs_sb_info *sbi)
 	 * for cp pack we can have max 1020*504 orphan entries
 	 */
 	sbi->n_orphans = 0;
-	sbi->max_orphans = (sbi->blocks_per_seg - 2 - NR_CURSEG_TYPE)
+	sbi->max_orphans = (sbi->blocks_per_seg - F2FS_CP_PACKS -
-				* F2FS_ORPHANS_PER_BLOCK;
+			NR_CURSEG_TYPE) * F2FS_ORPHANS_PER_BLOCK;
 }
 
 int __init create_checkpoint_caches(void)

fs/f2fs/data.c

@@ -53,7 +53,7 @@ static void f2fs_write_end_io(struct bio *bio, int err)
 		struct page *page = bvec->bv_page;
 
 		if (unlikely(err)) {
-			SetPageError(page);
+			set_page_dirty(page);
 			set_bit(AS_EIO, &page->mapping->flags);
 			f2fs_stop_checkpoint(sbi);
 		}
@@ -691,7 +691,7 @@ get_next:
 			allocated = true;
 			blkaddr = dn.data_blkaddr;
 		}
-		/* Give more consecutive addresses for the read ahead */
+		/* Give more consecutive addresses for the readahead */
 		if (blkaddr == (bh_result->b_blocknr + ofs)) {
 			ofs++;
 			dn.ofs_in_node++;
@@ -739,7 +739,7 @@ static int f2fs_read_data_page(struct file *file, struct page *page)
 
 	trace_f2fs_readpage(page, DATA);
 
-	/* If the file has inline data, try to read it directlly */
+	/* If the file has inline data, try to read it directly */
 	if (f2fs_has_inline_data(inode))
 		ret = f2fs_read_inline_data(inode, page);
 	else
@@ -836,10 +836,19 @@ write:
 
 	/* Dentry blocks are controlled by checkpoint */
 	if (S_ISDIR(inode->i_mode)) {
+		if (unlikely(f2fs_cp_error(sbi)))
+			goto redirty_out;
 		err = do_write_data_page(page, &fio);
 		goto done;
 	}
 
+	/* we should bypass data pages to proceed the kworkder jobs */
+	if (unlikely(f2fs_cp_error(sbi))) {
+		SetPageError(page);
+		unlock_page(page);
+		return 0;
+	}
+
 	if (!wbc->for_reclaim)
 		need_balance_fs = true;
 	else if (has_not_enough_free_secs(sbi, 0))
@@ -927,7 +936,7 @@ static void f2fs_write_failed(struct address_space *mapping, loff_t to)
 
 	if (to > inode->i_size) {
 		truncate_pagecache(inode, inode->i_size);
-		truncate_blocks(inode, inode->i_size);
+		truncate_blocks(inode, inode->i_size, true);
 	}
 }
 
@@ -946,7 +955,7 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 
 	f2fs_balance_fs(sbi);
 repeat:
-	err = f2fs_convert_inline_data(inode, pos + len);
+	err = f2fs_convert_inline_data(inode, pos + len, NULL);
 	if (err)
 		goto fail;
 

fs/f2fs/debug.c

@@ -32,7 +32,7 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	struct f2fs_stat_info *si = F2FS_STAT(sbi);
 	int i;
 
-	/* valid check of the segment numbers */
+	/* validation check of the segment numbers */
 	si->hit_ext = sbi->read_hit_ext;
 	si->total_ext = sbi->total_hit_ext;
 	si->ndirty_node = get_pages(sbi, F2FS_DIRTY_NODES);
@@ -152,7 +152,7 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	si->base_mem += NR_DIRTY_TYPE * f2fs_bitmap_size(TOTAL_SEGS(sbi));
 	si->base_mem += f2fs_bitmap_size(TOTAL_SECS(sbi));
 
-	/* buld nm */
+	/* build nm */
 	si->base_mem += sizeof(struct f2fs_nm_info);
 	si->base_mem += __bitmap_size(sbi, NAT_BITMAP);
 

fs/f2fs/dir.c

@@ -124,7 +124,7 @@ static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
 
 		/*
 		 * For the most part, it should be a bug when name_len is zero.
-		 * We stop here for figuring out where the bugs are occurred.
+		 * We stop here for figuring out where the bugs has occurred.
 		 */
 		f2fs_bug_on(!de->name_len);
 
@@ -391,7 +391,7 @@ put_error:
 error:
 	/* once the failed inode becomes a bad inode, i_mode is S_IFREG */
 	truncate_inode_pages(&inode->i_data, 0);
-	truncate_blocks(inode, 0);
+	truncate_blocks(inode, 0, false);
 	remove_dirty_dir_inode(inode);
 	remove_inode_page(inode);
 	return ERR_PTR(err);
@@ -563,7 +563,7 @@ fail:
 }
 
 /*
- * It only removes the dentry from the dentry page,corresponding name
+ * It only removes the dentry from the dentry page, corresponding name
  * entry in name page does not need to be touched during deletion.
  */
 void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,

fs/f2fs/f2fs.h

@@ -24,7 +24,7 @@
 #define f2fs_bug_on(condition)	BUG_ON(condition)
 #define f2fs_down_write(x, y)	down_write_nest_lock(x, y)
 #else
-#define f2fs_bug_on(condition)
+#define f2fs_bug_on(condition)	WARN_ON(condition)
 #define f2fs_down_write(x, y)	down_write(x)
 #endif
 
@@ -395,7 +395,7 @@ enum count_type {
 };
 
 /*
- * The below are the page types of bios used in submti_bio().
+ * The below are the page types of bios used in submit_bio().
  * The available types are:
  * DATA			User data pages. It operates as async mode.
  * NODE			Node pages. It operates as async mode.
@@ -470,7 +470,7 @@ struct f2fs_sb_info {
 	struct list_head dir_inode_list;	/* dir inode list */
 	spinlock_t dir_inode_lock;		/* for dir inode list lock */
 
-	/* basic file system units */
+	/* basic filesystem units */
 	unsigned int log_sectors_per_block;	/* log2 sectors per block */
 	unsigned int log_blocksize;		/* log2 block size */
 	unsigned int blocksize;			/* block size */
@@ -799,7 +799,7 @@ static inline block_t __start_cp_addr(struct f2fs_sb_info *sbi)
 
 	/*
 	 * odd numbered checkpoint should at cp segment 0
-	 * and even segent must be at cp segment 1
+	 * and even segment must be at cp segment 1
 	 */
 	if (!(ckpt_version & 1))
 		start_addr += sbi->blocks_per_seg;
@@ -1096,6 +1096,11 @@ static inline int f2fs_readonly(struct super_block *sb)
 	return sb->s_flags & MS_RDONLY;
 }
 
+static inline bool f2fs_cp_error(struct f2fs_sb_info *sbi)
+{
+	return is_set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
+}
+
 static inline void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi)
 {
 	set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
@@ -1117,7 +1122,7 @@ static inline void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi)
  */
 int f2fs_sync_file(struct file *, loff_t, loff_t, int);
 void truncate_data_blocks(struct dnode_of_data *);
-int truncate_blocks(struct inode *, u64);
+int truncate_blocks(struct inode *, u64, bool);
 void f2fs_truncate(struct inode *);
 int f2fs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 int f2fs_setattr(struct dentry *, struct iattr *);
@@ -1202,10 +1207,8 @@ int sync_node_pages(struct f2fs_sb_info *, nid_t, struct writeback_control *);
 bool alloc_nid(struct f2fs_sb_info *, nid_t *);
 void alloc_nid_done(struct f2fs_sb_info *, nid_t);
 void alloc_nid_failed(struct f2fs_sb_info *, nid_t);
-void recover_node_page(struct f2fs_sb_info *, struct page *,
-		struct f2fs_summary *, struct node_info *, block_t);
 void recover_inline_xattr(struct inode *, struct page *);
-bool recover_xattr_data(struct inode *, struct page *, block_t);
+void recover_xattr_data(struct inode *, struct page *, block_t);
 int recover_inode_page(struct f2fs_sb_info *, struct page *);
 int restore_node_summary(struct f2fs_sb_info *, unsigned int,
 				struct f2fs_summary_block *);
@@ -1238,8 +1241,6 @@ void write_data_page(struct page *, struct dnode_of_data *, block_t *,
 void rewrite_data_page(struct page *, block_t, struct f2fs_io_info *);
 void recover_data_page(struct f2fs_sb_info *, struct page *,
 				struct f2fs_summary *, block_t, block_t);
-void rewrite_node_page(struct f2fs_sb_info *, struct page *,
-				struct f2fs_summary *, block_t, block_t);
 void allocate_data_block(struct f2fs_sb_info *, struct page *,
 		block_t, block_t *, struct f2fs_summary *, int);
 void f2fs_wait_on_page_writeback(struct page *, enum page_type);
@@ -1262,6 +1263,7 @@ int ra_meta_pages(struct f2fs_sb_info *, int, int, int);
 long sync_meta_pages(struct f2fs_sb_info *, enum page_type, long);
 void add_dirty_inode(struct f2fs_sb_info *, nid_t, int type);
 void remove_dirty_inode(struct f2fs_sb_info *, nid_t, int type);
+void release_dirty_inode(struct f2fs_sb_info *);
 bool exist_written_data(struct f2fs_sb_info *, nid_t, int);
 int acquire_orphan_inode(struct f2fs_sb_info *);
 void release_orphan_inode(struct f2fs_sb_info *);
@@ -1439,8 +1441,8 @@ extern const struct inode_operations f2fs_special_inode_operations;
  */
 bool f2fs_may_inline(struct inode *);
 int f2fs_read_inline_data(struct inode *, struct page *);
-int f2fs_convert_inline_data(struct inode *, pgoff_t);
+int f2fs_convert_inline_data(struct inode *, pgoff_t, struct page *);
 int f2fs_write_inline_data(struct inode *, struct page *, unsigned int);
 void truncate_inline_data(struct inode *, u64);
-int recover_inline_data(struct inode *, struct page *);
+bool recover_inline_data(struct inode *, struct page *);
 #endif

fs/f2fs/file.c

@@ -41,6 +41,11 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 
 	sb_start_pagefault(inode->i_sb);
 
+	/* force to convert with normal data indices */
+	err = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, page);
+	if (err)
+		goto out;
+
 	/* block allocation */
 	f2fs_lock_op(sbi);
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
@@ -110,6 +115,25 @@ static int get_parent_ino(struct inode *inode, nid_t *pino)
 	return 1;
 }
 
+static inline bool need_do_checkpoint(struct inode *inode)
+{
+	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
+	bool need_cp = false;
+
+	if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1)
+		need_cp = true;
+	else if (file_wrong_pino(inode))
+		need_cp = true;
+	else if (!space_for_roll_forward(sbi))
+		need_cp = true;
+	else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
+		need_cp = true;
+	else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi)))
+		need_cp = true;
+
+	return need_cp;
+}
+
 int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
@@ -154,23 +178,12 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	/* guarantee free sections for fsync */
 	f2fs_balance_fs(sbi);
 
-	down_read(&fi->i_sem);
-
 	/*
 	 * Both of fdatasync() and fsync() are able to be recovered from
 	 * sudden-power-off.
 	 */
-	if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1)
+	down_read(&fi->i_sem);
-		need_cp = true;
+	need_cp = need_do_checkpoint(inode);
-	else if (file_wrong_pino(inode))
-		need_cp = true;
-	else if (!space_for_roll_forward(sbi))
-		need_cp = true;
-	else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
-		need_cp = true;
-	else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi)))
-		need_cp = true;
-
 	up_read(&fi->i_sem);
 
 	if (need_cp) {
@@ -288,7 +301,7 @@ static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
 		if (err && err != -ENOENT) {
 			goto fail;
 		} else if (err == -ENOENT) {
-			/* direct node is not exist */
+			/* direct node does not exists */
 			if (whence == SEEK_DATA) {
 				pgofs = PGOFS_OF_NEXT_DNODE(pgofs,
 							F2FS_I(inode));
@@ -417,7 +430,7 @@ out:
 	f2fs_put_page(page, 1);
 }
 
-int truncate_blocks(struct inode *inode, u64 from)
+int truncate_blocks(struct inode *inode, u64 from, bool lock)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	unsigned int blocksize = inode->i_sb->s_blocksize;
@@ -433,6 +446,7 @@ int truncate_blocks(struct inode *inode, u64 from)
 	free_from = (pgoff_t)
 			((from + blocksize - 1) >> (sbi->log_blocksize));
 
+	if (lock)
 		f2fs_lock_op(sbi);
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
@@ -440,6 +454,7 @@ int truncate_blocks(struct inode *inode, u64 from)
 	if (err) {
 		if (err == -ENOENT)
 			goto free_next;
+		if (lock)
 			f2fs_unlock_op(sbi);
 		trace_f2fs_truncate_blocks_exit(inode, err);
 		return err;
@@ -458,6 +473,7 @@ int truncate_blocks(struct inode *inode, u64 from)
 	f2fs_put_dnode(&dn);
 free_next:
 	err = truncate_inode_blocks(inode, free_from);
+	if (lock)
 		f2fs_unlock_op(sbi);
 done:
 	/* lastly zero out the first data page */
@@ -475,7 +491,7 @@ void f2fs_truncate(struct inode *inode)
 
 	trace_f2fs_truncate(inode);
 
-	if (!truncate_blocks(inode, i_size_read(inode))) {
+	if (!truncate_blocks(inode, i_size_read(inode), true)) {
 		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		mark_inode_dirty(inode);
 	}
@@ -533,7 +549,7 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
 
 	if ((attr->ia_valid & ATTR_SIZE) &&
 			attr->ia_size != i_size_read(inode)) {
-		err = f2fs_convert_inline_data(inode, attr->ia_size);
+		err = f2fs_convert_inline_data(inode, attr->ia_size, NULL);
 		if (err)
 			return err;
 
@@ -622,7 +638,7 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	loff_t off_start, off_end;
 	int ret = 0;
 
-	ret = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1);
+	ret = f2fs_convert_inline_data(inode, MAX_INLINE_DATA + 1, NULL);
 	if (ret)
 		return ret;
 
@@ -678,7 +694,7 @@ static int expand_inode_data(struct inode *inode, loff_t offset,
 	if (ret)
 		return ret;
 
-	ret = f2fs_convert_inline_data(inode, offset + len);
+	ret = f2fs_convert_inline_data(inode, offset + len, NULL);
 	if (ret)
 		return ret;
 

fs/f2fs/gc.c

@@ -58,7 +58,7 @@ static int gc_thread_func(void *data)
 		 * 3. IO subsystem is idle by checking the # of requests in
 		 *    bdev's request list.
 		 *
-		 * Note) We have to avoid triggering GCs too much frequently.
+		 * Note) We have to avoid triggering GCs frequently.
 		 * Because it is possible that some segments can be
 		 * invalidated soon after by user update or deletion.
 		 * So, I'd like to wait some time to collect dirty segments.
@@ -222,7 +222,7 @@ static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 
 	u = (vblocks * 100) >> sbi->log_blocks_per_seg;
 
-	/* Handle if the system time is changed by user */
+	/* Handle if the system time has changed by the user */
 	if (mtime < sit_i->min_mtime)
 		sit_i->min_mtime = mtime;
 	if (mtime > sit_i->max_mtime)
@@ -593,7 +593,7 @@ next_step:
 
 		if (phase == 2) {
 			inode = f2fs_iget(sb, dni.ino);
-			if (IS_ERR(inode))
+			if (IS_ERR(inode) || is_bad_inode(inode))
 				continue;
 
 			start_bidx = start_bidx_of_node(nofs, F2FS_I(inode));
@@ -693,7 +693,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi)
 gc_more:
 	if (unlikely(!(sbi->sb->s_flags & MS_ACTIVE)))
 		goto stop;
-	if (unlikely(is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ERROR_FLAG)))
+	if (unlikely(f2fs_cp_error(sbi)))
 		goto stop;
 
 	if (gc_type == BG_GC && has_not_enough_free_secs(sbi, nfree)) {

fs/f2fs/gc.h

@@ -91,7 +91,7 @@ static inline bool has_enough_invalid_blocks(struct f2fs_sb_info *sbi)
 	block_t invalid_user_blocks = sbi->user_block_count -
 					written_block_count(sbi);
 	/*
-	 * Background GC is triggered with the following condition.
+	 * Background GC is triggered with the following conditions.
 	 * 1. There are a number of invalid blocks.
 	 * 2. There is not enough free space.
 	 */

fs/f2fs/hash.c

@@ -42,7 +42,8 @@ static void TEA_transform(unsigned int buf[4], unsigned int const in[])
 	buf[1] += b1;
 }
 
-static void str2hashbuf(const char *msg, size_t len, unsigned int *buf, int num)
+static void str2hashbuf(const unsigned char *msg, size_t len,
+				unsigned int *buf, int num)
 {
 	unsigned pad, val;
 	int i;
@@ -73,9 +74,9 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info)
 {
 	__u32 hash;
 	f2fs_hash_t f2fs_hash;
-	const char *p;
+	const unsigned char *p;
 	__u32 in[8], buf[4];
-	const char *name = name_info->name;
+	const unsigned char *name = name_info->name;
 	size_t len = name_info->len;
 
 	if ((len <= 2) && (name[0] == '.') &&

fs/f2fs/inline.c

@@ -68,7 +68,7 @@ out:
 
 static int __f2fs_convert_inline_data(struct inode *inode, struct page *page)
 {
-	int err;
+	int err = 0;
 	struct page *ipage;
 	struct dnode_of_data dn;
 	void *src_addr, *dst_addr;
@@ -86,6 +86,10 @@ static int __f2fs_convert_inline_data(struct inode *inode, struct page *page)
 		goto out;
 	}
 
+	/* someone else converted inline_data already */
+	if (!f2fs_has_inline_data(inode))
+		goto out;
+
 	/*
 	 * i_addr[0] is not used for inline data,
 	 * so reserving new block will not destroy inline data
@@ -124,9 +128,10 @@ out:
 	return err;
 }
 
-int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size)
+int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size,
+						struct page *page)
 {
-	struct page *page;
+	struct page *new_page = page;
 	int err;
 
 	if (!f2fs_has_inline_data(inode))
@@ -134,12 +139,15 @@ int f2fs_convert_inline_data(struct inode *inode, pgoff_t to_size)
 	else if (to_size <= MAX_INLINE_DATA)
 		return 0;
 
-	page = grab_cache_page(inode->i_mapping, 0);
+	if (!page || page->index != 0) {
-	if (!page)
+		new_page = grab_cache_page(inode->i_mapping, 0);
+		if (!new_page)
 			return -ENOMEM;
+	}
 
-	err = __f2fs_convert_inline_data(inode, page);
+	err = __f2fs_convert_inline_data(inode, new_page);
-	f2fs_put_page(page, 1);
+	if (!page || page->index != 0)
+		f2fs_put_page(new_page, 1);
 	return err;
 }
 
@@ -199,7 +207,7 @@ void truncate_inline_data(struct inode *inode, u64 from)
 	f2fs_put_page(ipage, 1);
 }
 
-int recover_inline_data(struct inode *inode, struct page *npage)
+bool recover_inline_data(struct inode *inode, struct page *npage)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	struct f2fs_inode *ri = NULL;
@@ -218,7 +226,7 @@ int recover_inline_data(struct inode *inode, struct page *npage)
 		ri = F2FS_INODE(npage);
 
 	if (f2fs_has_inline_data(inode) &&
-			ri && ri->i_inline & F2FS_INLINE_DATA) {
+			ri && (ri->i_inline & F2FS_INLINE_DATA)) {
 process_inline:
 		ipage = get_node_page(sbi, inode->i_ino);
 		f2fs_bug_on(IS_ERR(ipage));
@@ -230,7 +238,7 @@ process_inline:
 		memcpy(dst_addr, src_addr, MAX_INLINE_DATA);
 		update_inode(inode, ipage);
 		f2fs_put_page(ipage, 1);
-		return -1;
+		return true;
 	}
 
 	if (f2fs_has_inline_data(inode)) {
@@ -242,10 +250,10 @@ process_inline:
 		clear_inode_flag(F2FS_I(inode), FI_INLINE_DATA);
 		update_inode(inode, ipage);
 		f2fs_put_page(ipage, 1);
-	} else if (ri && ri->i_inline & F2FS_INLINE_DATA) {
+	} else if (ri && (ri->i_inline & F2FS_INLINE_DATA)) {
-		truncate_blocks(inode, 0);
+		truncate_blocks(inode, 0, false);
 		set_inode_flag(F2FS_I(inode), FI_INLINE_DATA);
 		goto process_inline;
 	}
-	return 0;
+	return false;
 }

fs/f2fs/namei.c

@@ -134,9 +134,7 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	return 0;
 out:
 	clear_nlink(inode);
-	unlock_new_inode(inode);
+	iget_failed(inode);
-	make_bad_inode(inode);
-	iput(inode);
 	alloc_nid_failed(sbi, ino);
 	return err;
 }
@@ -267,9 +265,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 	return err;
 out:
 	clear_nlink(inode);
-	unlock_new_inode(inode);
+	iget_failed(inode);
-	make_bad_inode(inode);
-	iput(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -308,9 +304,7 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 out_fail:
 	clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
 	clear_nlink(inode);
-	unlock_new_inode(inode);
+	iget_failed(inode);
-	make_bad_inode(inode);
-	iput(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -354,9 +348,7 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 	return 0;
 out:
 	clear_nlink(inode);
-	unlock_new_inode(inode);
+	iget_failed(inode);
-	make_bad_inode(inode);
-	iput(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -688,9 +680,7 @@ release_out:
 out:
 	f2fs_unlock_op(sbi);
 	clear_nlink(inode);
-	unlock_new_inode(inode);
+	iget_failed(inode);
-	make_bad_inode(inode);
-	iput(inode);
 	alloc_nid_failed(sbi, inode->i_ino);
 	return err;
 }
@@ -704,7 +694,6 @@ const struct inode_operations f2fs_dir_inode_operations = {
 	.mkdir		= f2fs_mkdir,
 	.rmdir		= f2fs_rmdir,
 	.mknod		= f2fs_mknod,
-	.rename		= f2fs_rename,
 	.rename2	= f2fs_rename2,
 	.tmpfile	= f2fs_tmpfile,
 	.getattr	= f2fs_getattr,

fs/f2fs/node.c

@@ -237,7 +237,7 @@ retry:
 			nat_get_blkaddr(e) != NULL_ADDR &&
 			new_blkaddr == NEW_ADDR);
 
-	/* increament version no as node is removed */
+	/* increment version no as node is removed */
 	if (nat_get_blkaddr(e) != NEW_ADDR && new_blkaddr == NULL_ADDR) {
 		unsigned char version = nat_get_version(e);
 		nat_set_version(e, inc_node_version(version));
@@ -274,7 +274,7 @@ int try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink)
 }
 
 /*
- * This function returns always success
+ * This function always returns success
  */
 void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
 {
@@ -650,7 +650,7 @@ static int truncate_partial_nodes(struct dnode_of_data *dn,
 
 	/* get indirect nodes in the path */
 	for (i = 0; i < idx + 1; i++) {
-		/* refernece count'll be increased */
+		/* reference count'll be increased */
 		pages[i] = get_node_page(sbi, nid[i]);
 		if (IS_ERR(pages[i])) {
 			err = PTR_ERR(pages[i]);
@@ -823,22 +823,26 @@ int truncate_xattr_node(struct inode *inode, struct page *page)
  */
 void remove_inode_page(struct inode *inode)
 {
-	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
-	struct page *page;
-	nid_t ino = inode->i_ino;
 	struct dnode_of_data dn;
 
-	page = get_node_page(sbi, ino);
+	set_new_dnode(&dn, inode, NULL, NULL, inode->i_ino);
-	if (IS_ERR(page))
+	if (get_dnode_of_data(&dn, 0, LOOKUP_NODE))
 		return;
 
-	if (truncate_xattr_node(inode, page)) {
+	if (truncate_xattr_node(inode, dn.inode_page)) {
-		f2fs_put_page(page, 1);
+		f2fs_put_dnode(&dn);
 		return;
 	}
-	/* 0 is possible, after f2fs_new_inode() is failed */
+
+	/* remove potential inline_data blocks */
+	if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
+				S_ISLNK(inode->i_mode))
+		truncate_data_blocks_range(&dn, 1);
+
+	/* 0 is possible, after f2fs_new_inode() has failed */
 	f2fs_bug_on(inode->i_blocks != 0 && inode->i_blocks != 1);
-	set_new_dnode(&dn, inode, page, page, ino);
+
+	/* will put inode & node pages */
 	truncate_node(&dn);
 }
 
@@ -1129,7 +1133,10 @@ continue_unlock:
 				set_fsync_mark(page, 0);
 				set_dentry_mark(page, 0);
 			}
-			NODE_MAPPING(sbi)->a_ops->writepage(page, wbc);
+
+			if (NODE_MAPPING(sbi)->a_ops->writepage(page, wbc))
+				unlock_page(page);
+			else
 				wrote++;
 
 			if (--wbc->nr_to_write == 0)
@@ -1212,6 +1219,8 @@ static int f2fs_write_node_page(struct page *page,
 
 	if (unlikely(sbi->por_doing))
 		goto redirty_out;
+	if (unlikely(f2fs_cp_error(sbi)))
+		goto redirty_out;
 
 	f2fs_wait_on_page_writeback(page, NODE);
 
@@ -1540,15 +1549,6 @@ void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid)
 		kmem_cache_free(free_nid_slab, i);
 }
 
-void recover_node_page(struct f2fs_sb_info *sbi, struct page *page,
-		struct f2fs_summary *sum, struct node_info *ni,
-		block_t new_blkaddr)
-{
-	rewrite_node_page(sbi, page, sum, ni->blk_addr, new_blkaddr);
-	set_node_addr(sbi, ni, new_blkaddr, false);
-	clear_node_page_dirty(page);
-}
-
 void recover_inline_xattr(struct inode *inode, struct page *page)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
@@ -1557,40 +1557,33 @@ void recover_inline_xattr(struct inode *inode, struct page *page)
 	struct page *ipage;
 	struct f2fs_inode *ri;
 
-	if (!f2fs_has_inline_xattr(inode))
-		return;
-
-	if (!IS_INODE(page))
-		return;
-
-	ri = F2FS_INODE(page);
-	if (!(ri->i_inline & F2FS_INLINE_XATTR))
-		return;
-
 	ipage = get_node_page(sbi, inode->i_ino);
 	f2fs_bug_on(IS_ERR(ipage));
 
+	ri = F2FS_INODE(page);
+	if (!(ri->i_inline & F2FS_INLINE_XATTR)) {
+		clear_inode_flag(F2FS_I(inode), FI_INLINE_XATTR);
+		goto update_inode;
+	}
+
 	dst_addr = inline_xattr_addr(ipage);
 	src_addr = inline_xattr_addr(page);
 	inline_size = inline_xattr_size(inode);
 
 	f2fs_wait_on_page_writeback(ipage, NODE);
 	memcpy(dst_addr, src_addr, inline_size);
-
+update_inode:
 	update_inode(inode, ipage);
 	f2fs_put_page(ipage, 1);
 }
 
-bool recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
+void recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb);
 	nid_t prev_xnid = F2FS_I(inode)->i_xattr_nid;
 	nid_t new_xnid = nid_of_node(page);
 	struct node_info ni;
 
-	if (!f2fs_has_xattr_block(ofs_of_node(page)))
-		return false;
-
 	/* 1: invalidate the previous xattr nid */
 	if (!prev_xnid)
 		goto recover_xnid;
@@ -1618,7 +1611,6 @@ recover_xnid:
 	set_node_addr(sbi, &ni, blkaddr, false);
 
 	update_inode_page(inode);
-	return true;
 }
 
 int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
@@ -1651,6 +1643,7 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 	dst->i_blocks = cpu_to_le64(1);
 	dst->i_links = cpu_to_le32(1);
 	dst->i_xattr_nid = 0;
+	dst->i_inline = src->i_inline & F2FS_INLINE_XATTR;
 
 	new_ni = old_ni;
 	new_ni.ino = ino;
@@ -1659,13 +1652,14 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 		WARN_ON(1);
 	set_node_addr(sbi, &new_ni, NEW_ADDR, false);
 	inc_valid_inode_count(sbi);
+	set_page_dirty(ipage);
 	f2fs_put_page(ipage, 1);
 	return 0;
 }
 
 /*
  * ra_sum_pages() merge contiguous pages into one bio and submit.
- * these pre-readed pages are alloced in bd_inode's mapping tree.
+ * these pre-read pages are allocated in bd_inode's mapping tree.
  */
 static int ra_sum_pages(struct f2fs_sb_info *sbi, struct page **pages,
 				int start, int nrpages)
@@ -1709,7 +1703,7 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
 	for (i = 0; !err && i < last_offset; i += nrpages, addr += nrpages) {
 		nrpages = min(last_offset - i, bio_blocks);
 
-		/* read ahead node pages */
+		/* readahead node pages */
 		nrpages = ra_sum_pages(sbi, pages, addr, nrpages);
 		if (!nrpages)
 			return -ENOMEM;
@@ -1967,7 +1961,7 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
 	nm_i->max_nid = NAT_ENTRY_PER_BLOCK * nat_blocks;
 
 	/* not used nids: 0, node, meta, (and root counted as valid node) */
-	nm_i->available_nids = nm_i->max_nid - 3;
+	nm_i->available_nids = nm_i->max_nid - F2FS_RESERVED_NODE_NUM;
 	nm_i->fcnt = 0;
 	nm_i->nat_cnt = 0;
 	nm_i->ram_thresh = DEF_RAM_THRESHOLD;

fs/f2fs/recovery.c

@@ -62,8 +62,10 @@ static int recover_dentry(struct page *ipage, struct inode *inode)
 	}
 retry:
 	de = f2fs_find_entry(dir, &name, &page);
-	if (de && inode->i_ino == le32_to_cpu(de->ino))
+	if (de && inode->i_ino == le32_to_cpu(de->ino)) {
+		clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
 		goto out_unmap_put;
+	}
 	if (de) {
 		einode = f2fs_iget(inode->i_sb, le32_to_cpu(de->ino));
 		if (IS_ERR(einode)) {
@@ -300,14 +302,19 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 	struct node_info ni;
 	int err = 0, recovered = 0;
 
+	/* step 1: recover xattr */
+	if (IS_INODE(page)) {
 		recover_inline_xattr(inode, page);
+	} else if (f2fs_has_xattr_block(ofs_of_node(page))) {
+		recover_xattr_data(inode, page, blkaddr);
+		goto out;
+	}
 
+	/* step 2: recover inline data */
 	if (recover_inline_data(inode, page))
 		goto out;
 
-	if (recover_xattr_data(inode, page, blkaddr))
+	/* step 3: recover data indices */
-		goto out;
-
 	start = start_bidx_of_node(ofs_of_node(page), fi);
 	end = start + ADDRS_PER_PAGE(page, fi);
 
@@ -364,8 +371,6 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 	fill_node_footer(dn.node_page, dn.nid, ni.ino,
 					ofs_of_node(page), false);
 	set_page_dirty(dn.node_page);
-
-	recover_node_page(sbi, dn.node_page, &sum, &ni, blkaddr);
 err:
 	f2fs_put_dnode(&dn);
 	f2fs_unlock_op(sbi);
@@ -452,6 +457,9 @@ int recover_fsync_data(struct f2fs_sb_info *sbi)
 	/* step #1: find fsynced inode numbers */
 	sbi->por_doing = true;
 
+	/* prevent checkpoint */
+	mutex_lock(&sbi->cp_mutex);
+
 	blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
 
 	err = find_fsync_dnodes(sbi, &inode_list);
@@ -465,6 +473,7 @@ int recover_fsync_data(struct f2fs_sb_info *sbi)
 
 	/* step #2: recover data */
 	err = recover_data(sbi, &inode_list, CURSEG_WARM_NODE);
+	if (!err)
 		f2fs_bug_on(!list_empty(&inode_list));
 out:
 	destroy_fsync_dnodes(&inode_list);
@@ -482,8 +491,13 @@ out:
 		/* Flush all the NAT/SIT pages */
 		while (get_pages(sbi, F2FS_DIRTY_META))
 			sync_meta_pages(sbi, META, LONG_MAX);
+		set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
+		mutex_unlock(&sbi->cp_mutex);
 	} else if (need_writecp) {
+		mutex_unlock(&sbi->cp_mutex);
 		write_checkpoint(sbi, false);
+	} else {
+		mutex_unlock(&sbi->cp_mutex);
 	}
 	return err;
 }

fs/f2fs/segment.c

@@ -62,7 +62,7 @@ static inline unsigned long __reverse_ffs(unsigned long word)
 }
 
 /*
- * __find_rev_next(_zero)_bit is copied from lib/find_next_bit.c becasue
+ * __find_rev_next(_zero)_bit is copied from lib/find_next_bit.c because
  * f2fs_set_bit makes MSB and LSB reversed in a byte.
  * Example:
  *                             LSB <--> MSB
@@ -808,7 +808,7 @@ static void __refresh_next_blkoff(struct f2fs_sb_info *sbi,
 }
 
 /*
- * This function always allocates a used segment (from dirty seglist) by SSR
+ * This function always allocates a used segment(from dirty seglist) by SSR
  * manner, so it should recover the existing segment information of valid blocks
  */
 static void change_curseg(struct f2fs_sb_info *sbi, int type, bool reuse)
@@ -1103,55 +1103,6 @@ void recover_data_page(struct f2fs_sb_info *sbi,
 	mutex_unlock(&curseg->curseg_mutex);
 }
 
-void rewrite_node_page(struct f2fs_sb_info *sbi,
-			struct page *page, struct f2fs_summary *sum,
-			block_t old_blkaddr, block_t new_blkaddr)
-{
-	struct sit_info *sit_i = SIT_I(sbi);
-	int type = CURSEG_WARM_NODE;
-	struct curseg_info *curseg;
-	unsigned int segno, old_cursegno;
-	block_t next_blkaddr = next_blkaddr_of_node(page);
-	unsigned int next_segno = GET_SEGNO(sbi, next_blkaddr);
-	struct f2fs_io_info fio = {
-		.type = NODE,
-		.rw = WRITE_SYNC,
-	};
-
-	curseg = CURSEG_I(sbi, type);
-
-	mutex_lock(&curseg->curseg_mutex);
-	mutex_lock(&sit_i->sentry_lock);
-
-	segno = GET_SEGNO(sbi, new_blkaddr);
-	old_cursegno = curseg->segno;
-
-	/* change the current segment */
-	if (segno != curseg->segno) {
-		curseg->next_segno = segno;
-		change_curseg(sbi, type, true);
-	}
-	curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, new_blkaddr);
-	__add_sum_entry(sbi, type, sum);
-
-	/* change the current log to the next block addr in advance */
-	if (next_segno != segno) {
-		curseg->next_segno = next_segno;
-		change_curseg(sbi, type, true);
-	}
-	curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, next_blkaddr);
-
-	/* rewrite node page */
-	set_page_writeback(page);
-	f2fs_submit_page_mbio(sbi, page, new_blkaddr, &fio);
-	f2fs_submit_merged_bio(sbi, NODE, WRITE);
-	refresh_sit_entry(sbi, old_blkaddr, new_blkaddr);
-	locate_dirty_segment(sbi, old_cursegno);
-
-	mutex_unlock(&sit_i->sentry_lock);
-	mutex_unlock(&curseg->curseg_mutex);
-}
-
 static inline bool is_merged_page(struct f2fs_sb_info *sbi,
 					struct page *page, enum page_type type)
 {

fs/f2fs/segment.h

@@ -549,7 +549,7 @@ static inline void verify_block_addr(struct f2fs_sb_info *sbi, block_t blk_addr)
 }
 
 /*
- * Summary block is always treated as invalid block
+ * Summary block is always treated as an invalid block
  */
 static inline void check_block_count(struct f2fs_sb_info *sbi,
 		int segno, struct f2fs_sit_entry *raw_sit)

fs/f2fs/super.c

@@ -432,9 +432,15 @@ static void f2fs_put_super(struct super_block *sb)
 	stop_gc_thread(sbi);
 
 	/* We don't need to do checkpoint when it's clean */
-	if (sbi->s_dirty && get_pages(sbi, F2FS_DIRTY_NODES))
+	if (sbi->s_dirty)
 		write_checkpoint(sbi, true);
 
+	/*
+	 * normally superblock is clean, so we need to release this.
+	 * In addition, EIO will skip do checkpoint, we need this as well.
+	 */
+	release_dirty_inode(sbi);
+
 	iput(sbi->node_inode);
 	iput(sbi->meta_inode);
 
@@ -457,9 +463,6 @@ int f2fs_sync_fs(struct super_block *sb, int sync)
 
 	trace_f2fs_sync_fs(sb, sync);
 
-	if (!sbi->s_dirty && !get_pages(sbi, F2FS_DIRTY_NODES))
-		return 0;
-
 	if (sync) {
 		mutex_lock(&sbi->gc_mutex);
 		write_checkpoint(sbi, false);
@@ -505,8 +508,8 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_bfree = buf->f_blocks - valid_user_blocks(sbi) - ovp_count;
 	buf->f_bavail = user_block_count - valid_user_blocks(sbi);
 
-	buf->f_files = sbi->total_node_count;
+	buf->f_files = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
-	buf->f_ffree = sbi->total_node_count - valid_inode_count(sbi);
+	buf->f_ffree = buf->f_files - valid_inode_count(sbi);
 
 	buf->f_namelen = F2FS_NAME_LEN;
 	buf->f_fsid.val[0] = (u32)id;
@@ -663,7 +666,7 @@ restore_gc:
 	if (need_restart_gc) {
 		if (start_gc_thread(sbi))
 			f2fs_msg(sbi->sb, KERN_WARNING,
-				"background gc thread is stop");
+				"background gc thread has stopped");
 	} else if (need_stop_gc) {
 		stop_gc_thread(sbi);
 	}
@@ -812,7 +815,7 @@ static int sanity_check_ckpt(struct f2fs_sb_info *sbi)
 	if (unlikely(fsmeta >= total))
 		return 1;
 
-	if (unlikely(is_set_ckpt_flags(ckpt, CP_ERROR_FLAG))) {
+	if (unlikely(f2fs_cp_error(sbi))) {
 		f2fs_msg(sbi->sb, KERN_ERR, "A bug case: need to run fsck");
 		return 1;
 	}
@@ -899,8 +902,10 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	struct buffer_head *raw_super_buf;
 	struct inode *root;
 	long err = -EINVAL;
+	bool retry = true;
 	int i;
 
+try_onemore:
 	/* allocate memory for f2fs-specific super block info */
 	sbi = kzalloc(sizeof(struct f2fs_sb_info), GFP_KERNEL);
 	if (!sbi)
@@ -1080,9 +1085,11 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	/* recover fsynced data */
 	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
 		err = recover_fsync_data(sbi);
-		if (err)
+		if (err) {
 			f2fs_msg(sb, KERN_ERR,
 				"Cannot recover all fsync data errno=%ld", err);
+			goto free_kobj;
+		}
 	}
 
 	/*
@@ -1123,6 +1130,13 @@ free_sb_buf:
 	brelse(raw_super_buf);
 free_sbi:
 	kfree(sbi);
+
+	/* give only one another chance */
+	if (retry) {
+		retry = 0;
+		shrink_dcache_sb(sb);
+		goto try_onemore;
+	}
 	return err;
 }
 

fs/f2fs/xattr.c

@@ -528,7 +528,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 		int free;
 		/*
 		 * If value is NULL, it is remove operation.
-		 * In case of update operation, we caculate free.
+		 * In case of update operation, we calculate free.
 		 */
 		free = MIN_OFFSET(inode) - ((char *)last - (char *)base_addr);
 		if (found)

include/linux/f2fs_fs.h

@@ -24,6 +24,9 @@
 #define NULL_ADDR		((block_t)0)	/* used as block_t addresses */
 #define NEW_ADDR		((block_t)-1)	/* used as block_t addresses */
 
+/* 0, 1(node nid), 2(meta nid) are reserved node id */
+#define F2FS_RESERVED_NODE_NUM		3
+
 #define F2FS_ROOT_INO(sbi)	(sbi->root_ino_num)
 #define F2FS_NODE_INO(sbi)	(sbi->node_ino_num)
 #define F2FS_META_INO(sbi)	(sbi->meta_ino_num)
@@ -87,6 +90,8 @@ struct f2fs_super_block {
 #define CP_ORPHAN_PRESENT_FLAG	0x00000002
 #define CP_UMOUNT_FLAG		0x00000001
 
+#define F2FS_CP_PACKS		2	/* # of checkpoint packs */
+
 struct f2fs_checkpoint {
 	__le64 checkpoint_ver;		/* checkpoint block version number */
 	__le64 user_block_count;	/* # of user blocks */
@@ -123,6 +128,9 @@ struct f2fs_checkpoint {
  */
 #define F2FS_ORPHANS_PER_BLOCK	1020
 
+#define GET_ORPHAN_BLOCKS(n)	((n + F2FS_ORPHANS_PER_BLOCK - 1) / \
+					F2FS_ORPHANS_PER_BLOCK)
+
 struct f2fs_orphan_block {
 	__le32 ino[F2FS_ORPHANS_PER_BLOCK];	/* inode numbers */
 	__le32 reserved;	/* reserved */
@@ -144,6 +152,7 @@ struct f2fs_extent {
 #define F2FS_NAME_LEN		255
 #define F2FS_INLINE_XATTR_ADDRS	50	/* 200 bytes for inline xattrs */
 #define DEF_ADDRS_PER_INODE	923	/* Address Pointers in an Inode */
+#define DEF_NIDS_PER_INODE	5	/* Node IDs in an Inode */
 #define ADDRS_PER_INODE(fi)	addrs_per_inode(fi)
 #define ADDRS_PER_BLOCK		1018	/* Address Pointers in a Direct Block */
 #define NIDS_PER_BLOCK		1018	/* Node IDs in an Indirect Block */
@@ -163,8 +172,9 @@ struct f2fs_extent {
 #define MAX_INLINE_DATA		(sizeof(__le32) * (DEF_ADDRS_PER_INODE - \
 						F2FS_INLINE_XATTR_ADDRS - 1))
 
-#define INLINE_DATA_OFFSET	(PAGE_CACHE_SIZE - sizeof(struct node_footer) \
+#define INLINE_DATA_OFFSET	(PAGE_CACHE_SIZE - sizeof(struct node_footer) -\
-			- sizeof(__le32) * (DEF_ADDRS_PER_INODE + 5 - 1))
+				sizeof(__le32) * (DEF_ADDRS_PER_INODE + \
+				DEF_NIDS_PER_INODE - 1))
 
 struct f2fs_inode {
 	__le16 i_mode;			/* file mode */
@@ -194,7 +204,7 @@ struct f2fs_inode {
 
 	__le32 i_addr[DEF_ADDRS_PER_INODE];	/* Pointers to data blocks */
 
-	__le32 i_nid[5];		/* direct(2), indirect(2),
+	__le32 i_nid[DEF_NIDS_PER_INODE];	/* direct(2), indirect(2),
 						double_indirect(1) node id */
 } __packed;