Merge tag 'for-f2fs-4.12' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "In this round, we've focused on enhancing performance with regards to
  block allocation, GC, and discard/in-place-update IO controls. There
  are a bunch of clean-ups as well as minor bug fixes.

  Enhancements:
   - disable heap-based allocation by default
   - issue small-sized discard commands by default
   - change the policy of data hotness for logging
   - distinguish IOs in terms of size and wbc type
   - start SSR earlier to avoid foreground GC
   - enhance data structures managing discard commands
   - enhance in-place update flow
   - add some more fault injection routines
   - secure one more xattr entry

  Bug fixes:
   - calculate victim cost for GC correctly
   - remain correct victim segment number for GC
   - race condition in nid allocator and initializer
   - stale pointer produced by atomic_writes
   - fix missing REQ_SYNC for flush commands
   - handle missing errors in more corner cases"

* tag 'for-f2fs-4.12' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (111 commits)
  f2fs: fix a mount fail for wrong next_scan_nid
  f2fs: enhance scalability of trace macro
  f2fs: relocate inode_{,un}lock in F2FS_IOC_SETFLAGS
  f2fs: Make flush bios explicitely sync
  f2fs: show available_nids in f2fs/status
  f2fs: flush dirty nats periodically
  f2fs: introduce CP_TRIMMED_FLAG to avoid unneeded discard
  f2fs: allow cpc->reason to indicate more than one reason
  f2fs: release cp and dnode lock before IPU
  f2fs: shrink size of struct discard_cmd
  f2fs: don't hold cmd_lock during waiting discard command
  f2fs: nullify fio->encrypted_page for each writes
  f2fs: sanity check segment count
  f2fs: introduce valid_ipu_blkaddr to clean up
  f2fs: lookup extent cache first under IPU scenario
  f2fs: reconstruct code to write a data page
  f2fs: introduce __wait_discard_cmd
  f2fs: introduce __issue_discard_cmd
  f2fs: enable small discard by default
  f2fs: delay awaking discard thread
  ...

fs/f2fs/checkpoint.c

@@ -275,10 +275,11 @@ static int f2fs_write_meta_pages(struct address_space *mapping,
 		get_pages(sbi, F2FS_DIRTY_META) < nr_pages_to_skip(sbi, META))
 		goto skip_write;
 
-	trace_f2fs_writepages(mapping->host, wbc, META);
+	/* if locked failed, cp will flush dirty pages instead */
+	if (!mutex_trylock(&sbi->cp_mutex))
+		goto skip_write;
 
-	/* if mounting is failed, skip writing node pages */
-	mutex_lock(&sbi->cp_mutex);
+	trace_f2fs_writepages(mapping->host, wbc, META);
 	diff = nr_pages_to_write(sbi, META, wbc);
 	written = sync_meta_pages(sbi, META, wbc->nr_to_write);
 	mutex_unlock(&sbi->cp_mutex);
@@ -567,7 +568,7 @@ static int recover_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 	if (ni.blk_addr != NULL_ADDR) {
 		set_sbi_flag(sbi, SBI_NEED_FSCK);
 		f2fs_msg(sbi->sb, KERN_WARNING,
-			"%s: orphan failed (ino=%x), run fsck to fix.",
+			"%s: orphan failed (ino=%x) by kernel, retry mount.",
 				__func__, ino);
 		return -EIO;
 	}
@@ -677,7 +678,7 @@ static int get_checkpoint_version(struct f2fs_sb_info *sbi, block_t cp_addr,
 	*cp_block = (struct f2fs_checkpoint *)page_address(*cp_page);
 
 	crc_offset = le32_to_cpu((*cp_block)->checksum_offset);
-	if (crc_offset >= blk_size) {
+	if (crc_offset > (blk_size - sizeof(__le32))) {
 		f2fs_msg(sbi->sb, KERN_WARNING,
 			"invalid crc_offset: %zu", crc_offset);
 		return -EINVAL;
@@ -816,7 +817,9 @@ static void __add_dirty_inode(struct inode *inode, enum inode_type type)
 		return;
 
 	set_inode_flag(inode, flag);
-	list_add_tail(&F2FS_I(inode)->dirty_list, &sbi->inode_list[type]);
+	if (!f2fs_is_volatile_file(inode))
+		list_add_tail(&F2FS_I(inode)->dirty_list,
+						&sbi->inode_list[type]);
 	stat_inc_dirty_inode(sbi, type);
 }
 
@@ -941,6 +944,19 @@ int f2fs_sync_inode_meta(struct f2fs_sb_info *sbi)
 	return 0;
 }
 
+static void __prepare_cp_block(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	nid_t last_nid = nm_i->next_scan_nid;
+
+	next_free_nid(sbi, &last_nid);
+	ckpt->valid_block_count = cpu_to_le64(valid_user_blocks(sbi));
+	ckpt->valid_node_count = cpu_to_le32(valid_node_count(sbi));
+	ckpt->valid_inode_count = cpu_to_le32(valid_inode_count(sbi));
+	ckpt->next_free_nid = cpu_to_le32(last_nid);
+}
+
 /*
  * Freeze all the FS-operations for checkpoint.
  */
@@ -964,21 +980,26 @@ static int block_operations(struct f2fs_sb_info *sbi)
 		err = sync_dirty_inodes(sbi, DIR_INODE);
 		if (err)
 			goto out;
+		cond_resched();
 		goto retry_flush_dents;
 	}
 
+	/*
+	 * POR: we should ensure that there are no dirty node pages
+	 * until finishing nat/sit flush. inode->i_blocks can be updated.
+	 */
+	down_write(&sbi->node_change);
+
 	if (get_pages(sbi, F2FS_DIRTY_IMETA)) {
+		up_write(&sbi->node_change);
 		f2fs_unlock_all(sbi);
 		err = f2fs_sync_inode_meta(sbi);
 		if (err)
 			goto out;
+		cond_resched();
 		goto retry_flush_dents;
 	}
 
-	/*
-	 * POR: we should ensure that there are no dirty node pages
-	 * until finishing nat/sit flush.
-	 */
 retry_flush_nodes:
 	down_write(&sbi->node_write);
 
@@ -986,11 +1007,20 @@ static int block_operations(struct f2fs_sb_info *sbi)
 		up_write(&sbi->node_write);
 		err = sync_node_pages(sbi, &wbc);
 		if (err) {
+			up_write(&sbi->node_change);
 			f2fs_unlock_all(sbi);
 			goto out;
 		}
+		cond_resched();
 		goto retry_flush_nodes;
 	}
+
+	/*
+	 * sbi->node_change is used only for AIO write_begin path which produces
+	 * dirty node blocks and some checkpoint values by block allocation.
+	 */
+	__prepare_cp_block(sbi);
+	up_write(&sbi->node_change);
 out:
 	blk_finish_plug(&plug);
 	return err;
@@ -1024,16 +1054,20 @@ static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	spin_lock(&sbi->cp_lock);
 
-	if (cpc->reason == CP_UMOUNT && ckpt->cp_pack_total_block_count >
+	if ((cpc->reason & CP_UMOUNT) &&
+			le32_to_cpu(ckpt->cp_pack_total_block_count) >
 			sbi->blocks_per_seg - NM_I(sbi)->nat_bits_blocks)
 		disable_nat_bits(sbi, false);
 
-	if (cpc->reason == CP_UMOUNT)
+	if (cpc->reason & CP_TRIMMED)
+		__set_ckpt_flags(ckpt, CP_TRIMMED_FLAG);
+
+	if (cpc->reason & CP_UMOUNT)
 		__set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
 	else
 		__clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
 
-	if (cpc->reason == CP_FASTBOOT)
+	if (cpc->reason & CP_FASTBOOT)
 		__set_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
 	else
 		__clear_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
@@ -1057,7 +1091,6 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	unsigned long orphan_num = sbi->im[ORPHAN_INO].ino_num;
-	nid_t last_nid = nm_i->next_scan_nid;
 	block_t start_blk;
 	unsigned int data_sum_blocks, orphan_blocks;
 	__u32 crc32 = 0;
@@ -1074,14 +1107,11 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 			return -EIO;
 	}
 
-	next_free_nid(sbi, &last_nid);
-
 	/*
 	 * modify checkpoint
 	 * version number is already updated
 	 */
 	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 < NR_CURSEG_NODE_TYPE; i++) {
 		ckpt->cur_node_segno[i] =
@@ -1100,10 +1130,6 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 				curseg_alloc_type(sbi, i + CURSEG_HOT_DATA);
 	}
 
-	ckpt->valid_node_count = cpu_to_le32(valid_node_count(sbi));
-	ckpt->valid_inode_count = cpu_to_le32(valid_inode_count(sbi));
-	ckpt->next_free_nid = cpu_to_le32(last_nid);
-
 	/* 2 cp  + n data seg summary + orphan inode blocks */
 	data_sum_blocks = npages_for_summary_flush(sbi, false);
 	spin_lock(&sbi->cp_lock);
@@ -1143,7 +1169,6 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	/* write nat bits */
 	if (enabled_nat_bits(sbi, cpc)) {
 		__u64 cp_ver = cur_cp_version(ckpt);
-		unsigned int i;
 		block_t blk;
 
 		cp_ver |= ((__u64)crc32 << 32);
@@ -1250,8 +1275,8 @@ int write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	mutex_lock(&sbi->cp_mutex);
 
 	if (!is_sbi_flag_set(sbi, SBI_IS_DIRTY) &&
-		(cpc->reason == CP_FASTBOOT || cpc->reason == CP_SYNC ||
-		(cpc->reason == CP_DISCARD && !sbi->discard_blks)))
+		((cpc->reason & CP_FASTBOOT) || (cpc->reason & CP_SYNC) ||
+		((cpc->reason & CP_DISCARD) && !sbi->discard_blks)))
 		goto out;
 	if (unlikely(f2fs_cp_error(sbi))) {
 		err = -EIO;
@@ -1273,7 +1298,7 @@ int write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	f2fs_flush_merged_bios(sbi);
 
 	/* this is the case of multiple fstrims without any changes */
-	if (cpc->reason == CP_DISCARD) {
+	if (cpc->reason & CP_DISCARD) {
 		if (!exist_trim_candidates(sbi, cpc)) {
 			unblock_operations(sbi);
 			goto out;
@@ -1311,7 +1336,7 @@ int write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	unblock_operations(sbi);
 	stat_inc_cp_count(sbi->stat_info);
 
-	if (cpc->reason == CP_RECOVERY)
+	if (cpc->reason & CP_RECOVERY)
 		f2fs_msg(sbi->sb, KERN_NOTICE,
 			"checkpoint: version = %llx", ckpt_ver);
 

fs/f2fs/debug.c

@@ -51,15 +51,26 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->ndirty_all = sbi->ndirty_inode[DIRTY_META];
 	si->inmem_pages = get_pages(sbi, F2FS_INMEM_PAGES);
 	si->aw_cnt = atomic_read(&sbi->aw_cnt);
+	si->vw_cnt = atomic_read(&sbi->vw_cnt);
 	si->max_aw_cnt = atomic_read(&sbi->max_aw_cnt);
+	si->max_vw_cnt = atomic_read(&sbi->max_vw_cnt);
 	si->nr_wb_cp_data = get_pages(sbi, F2FS_WB_CP_DATA);
 	si->nr_wb_data = get_pages(sbi, F2FS_WB_DATA);
-	if (SM_I(sbi) && SM_I(sbi)->fcc_info)
-		si->nr_flush =
-			atomic_read(&SM_I(sbi)->fcc_info->submit_flush);
-	if (SM_I(sbi) && SM_I(sbi)->dcc_info)
-		si->nr_discard =
-			atomic_read(&SM_I(sbi)->dcc_info->submit_discard);
+	if (SM_I(sbi) && SM_I(sbi)->fcc_info) {
+		si->nr_flushed =
+			atomic_read(&SM_I(sbi)->fcc_info->issued_flush);
+		si->nr_flushing =
+			atomic_read(&SM_I(sbi)->fcc_info->issing_flush);
+	}
+	if (SM_I(sbi) && SM_I(sbi)->dcc_info) {
+		si->nr_discarded =
+			atomic_read(&SM_I(sbi)->dcc_info->issued_discard);
+		si->nr_discarding =
+			atomic_read(&SM_I(sbi)->dcc_info->issing_discard);
+		si->nr_discard_cmd =
+			atomic_read(&SM_I(sbi)->dcc_info->discard_cmd_cnt);
+		si->undiscard_blks = SM_I(sbi)->dcc_info->undiscard_blks;
+	}
 	si->total_count = (int)sbi->user_block_count / sbi->blocks_per_seg;
 	si->rsvd_segs = reserved_segments(sbi);
 	si->overp_segs = overprovision_segments(sbi);
@@ -86,6 +97,7 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->sits = MAIN_SEGS(sbi);
 	si->dirty_sits = SIT_I(sbi)->dirty_sentries;
 	si->free_nids = NM_I(sbi)->nid_cnt[FREE_NID_LIST];
+	si->avail_nids = NM_I(sbi)->available_nids;
 	si->alloc_nids = NM_I(sbi)->nid_cnt[ALLOC_NID_LIST];
 	si->bg_gc = sbi->bg_gc;
 	si->util_free = (int)(free_user_blocks(sbi) >> sbi->log_blocks_per_seg)
@@ -99,8 +111,8 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_NODE; i++) {
 		struct curseg_info *curseg = CURSEG_I(sbi, i);
 		si->curseg[i] = curseg->segno;
-		si->cursec[i] = curseg->segno / sbi->segs_per_sec;
-		si->curzone[i] = si->cursec[i] / sbi->secs_per_zone;
+		si->cursec[i] = GET_SEC_FROM_SEG(sbi, curseg->segno);
+		si->curzone[i] = GET_ZONE_FROM_SEC(sbi, si->cursec[i]);
 	}
 
 	for (i = 0; i < 2; i++) {
@@ -124,10 +136,10 @@ static void update_sit_info(struct f2fs_sb_info *sbi)
 
 	bimodal = 0;
 	total_vblocks = 0;
-	blks_per_sec = sbi->segs_per_sec * sbi->blocks_per_seg;
+	blks_per_sec = BLKS_PER_SEC(sbi);
 	hblks_per_sec = blks_per_sec / 2;
 	for (segno = 0; segno < MAIN_SEGS(sbi); segno += sbi->segs_per_sec) {
-		vblocks = get_valid_blocks(sbi, segno, sbi->segs_per_sec);
+		vblocks = get_valid_blocks(sbi, segno, true);
 		dist = abs(vblocks - hblks_per_sec);
 		bimodal += dist * dist;
 
@@ -156,7 +168,11 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	if (si->base_mem)
 		goto get_cache;
 
-	si->base_mem = sizeof(struct f2fs_sb_info) + sbi->sb->s_blocksize;
+	/* build stat */
+	si->base_mem = sizeof(struct f2fs_stat_info);
+
+	/* build superblock */
+	si->base_mem += sizeof(struct f2fs_sb_info) + sbi->sb->s_blocksize;
 	si->base_mem += 2 * sizeof(struct f2fs_inode_info);
 	si->base_mem += sizeof(*sbi->ckpt);
 	si->base_mem += sizeof(struct percpu_counter) * NR_COUNT_TYPE;
@@ -208,8 +224,11 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	/* build merge flush thread */
 	if (SM_I(sbi)->fcc_info)
 		si->cache_mem += sizeof(struct flush_cmd_control);
-	if (SM_I(sbi)->dcc_info)
+	if (SM_I(sbi)->dcc_info) {
 		si->cache_mem += sizeof(struct discard_cmd_control);
+		si->cache_mem += sizeof(struct discard_cmd) *
+			atomic_read(&SM_I(sbi)->dcc_info->discard_cmd_cnt);
+	}
 
 	/* free nids */
 	si->cache_mem += (NM_I(sbi)->nid_cnt[FREE_NID_LIST] +
@@ -330,11 +349,16 @@ static int stat_show(struct seq_file *s, void *v)
 		seq_printf(s, "  - Inner Struct Count: tree: %d(%d), node: %d\n",
 				si->ext_tree, si->zombie_tree, si->ext_node);
 		seq_puts(s, "\nBalancing F2FS Async:\n");
-		seq_printf(s, "  - IO (CP: %4d, Data: %4d, Flush: %4d, Discard: %4d)\n",
+		seq_printf(s, "  - IO (CP: %4d, Data: %4d, Flush: (%4d %4d), "
+			"Discard: (%4d %4d)) cmd: %4d undiscard:%4u\n",
 			   si->nr_wb_cp_data, si->nr_wb_data,
-			   si->nr_flush, si->nr_discard);
-		seq_printf(s, "  - inmem: %4d, atomic IO: %4d (Max. %4d)\n",
-			   si->inmem_pages, si->aw_cnt, si->max_aw_cnt);
+			   si->nr_flushing, si->nr_flushed,
+			   si->nr_discarding, si->nr_discarded,
+			   si->nr_discard_cmd, si->undiscard_blks);
+		seq_printf(s, "  - inmem: %4d, atomic IO: %4d (Max. %4d), "
+			"volatile IO: %4d (Max. %4d)\n",
+			   si->inmem_pages, si->aw_cnt, si->max_aw_cnt,
+			   si->vw_cnt, si->max_vw_cnt);
 		seq_printf(s, "  - nodes: %4d in %4d\n",
 			   si->ndirty_node, si->node_pages);
 		seq_printf(s, "  - dents: %4d in dirs:%4d (%4d)\n",
@@ -347,8 +371,8 @@ static int stat_show(struct seq_file *s, void *v)
 			   si->ndirty_imeta);
 		seq_printf(s, "  - NATs: %9d/%9d\n  - SITs: %9d/%9d\n",
 			   si->dirty_nats, si->nats, si->dirty_sits, si->sits);
-		seq_printf(s, "  - free_nids: %9d, alloc_nids: %9d\n",
-			   si->free_nids, si->alloc_nids);
+		seq_printf(s, "  - free_nids: %9d/%9d\n  - alloc_nids: %9d\n",
+			   si->free_nids, si->avail_nids, si->alloc_nids);
 		seq_puts(s, "\nDistribution of User Blocks:");
 		seq_puts(s, " [ valid | invalid | free ]\n");
 		seq_puts(s, "  [");
@@ -434,7 +458,9 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi)
 	atomic_set(&sbi->inplace_count, 0);
 
 	atomic_set(&sbi->aw_cnt, 0);
+	atomic_set(&sbi->vw_cnt, 0);
 	atomic_set(&sbi->max_aw_cnt, 0);
+	atomic_set(&sbi->max_vw_cnt, 0);
 
 	mutex_lock(&f2fs_stat_mutex);
 	list_add_tail(&si->stat_list, &f2fs_stat_list);

fs/f2fs/dir.c

@@ -94,7 +94,7 @@ static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
 
 	dentry_blk = (struct f2fs_dentry_block *)kmap(dentry_page);
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 1);
+	make_dentry_ptr_block(NULL, &d, dentry_blk);
 	de = find_target_dentry(fname, namehash, max_slots, &d);
 	if (de)
 		*res_page = dentry_page;
@@ -192,14 +192,10 @@ static struct f2fs_dir_entry *find_in_level(struct inode *dir,
 		f2fs_put_page(dentry_page, 0);
 	}
 
-	/* This is to increase the speed of f2fs_create */
-	if (!de && room) {
-		F2FS_I(dir)->task = current;
-		if (F2FS_I(dir)->chash != namehash) {
+	if (!de && room && F2FS_I(dir)->chash != namehash) {
 		F2FS_I(dir)->chash = namehash;
 		F2FS_I(dir)->clevel = level;
 	}
-	}
 
 	return de;
 }
@@ -239,6 +235,9 @@ struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir,
 			break;
 	}
 out:
+	/* This is to increase the speed of f2fs_create */
+	if (!de)
+		F2FS_I(dir)->task = current;
 	return de;
 }
 
@@ -322,24 +321,6 @@ static void init_dent_inode(const struct qstr *name, struct page *ipage)
 	set_page_dirty(ipage);
 }
 
-int update_dent_inode(struct inode *inode, struct inode *to,
-					const struct qstr *name)
-{
-	struct page *page;
-
-	if (file_enc_name(to))
-		return 0;
-
-	page = get_node_page(F2FS_I_SB(inode), inode->i_ino);
-	if (IS_ERR(page))
-		return PTR_ERR(page);
-
-	init_dent_inode(name, page);
-	f2fs_put_page(page, 1);
-
-	return 0;
-}
-
 void do_make_empty_dir(struct inode *inode, struct inode *parent,
 					struct f2fs_dentry_ptr *d)
 {
@@ -369,7 +350,7 @@ static int make_empty_dir(struct inode *inode,
 
 	dentry_blk = kmap_atomic(dentry_page);
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 1);
+	make_dentry_ptr_block(NULL, &d, dentry_blk);
 	do_make_empty_dir(inode, parent, &d);
 
 	kunmap_atomic(dentry_blk);
@@ -423,8 +404,11 @@ struct page *init_inode_metadata(struct inode *inode, struct inode *dir,
 		set_cold_node(inode, page);
 	}
 
-	if (new_name)
+	if (new_name) {
 		init_dent_inode(new_name, page);
+		if (f2fs_encrypted_inode(dir))
+			file_set_enc_name(inode);
+	}
 
 	/*
 	 * This file should be checkpointed during fsync.
@@ -584,11 +568,9 @@ int f2fs_add_regular_entry(struct inode *dir, const struct qstr *new_name,
 			err = PTR_ERR(page);
 			goto fail;
 		}
-		if (f2fs_encrypted_inode(dir))
-			file_set_enc_name(inode);
 	}
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 1);
+	make_dentry_ptr_block(NULL, &d, dentry_blk);
 	f2fs_update_dentry(ino, mode, &d, new_name, dentry_hash, bit_pos);
 
 	set_page_dirty(dentry_page);
@@ -896,7 +878,7 @@ static int f2fs_readdir(struct file *file, struct dir_context *ctx)
 
 		dentry_blk = kmap(dentry_page);
 
-		make_dentry_ptr(inode, &d, (void *)dentry_blk, 1);
+		make_dentry_ptr_block(inode, &d, dentry_blk);
 
 		err = f2fs_fill_dentries(ctx, &d,
 				n * NR_DENTRY_IN_BLOCK, &fstr);

fs/f2fs/extent_cache.c

@@ -18,6 +18,179 @@
 #include "node.h"
 #include <trace/events/f2fs.h>
 
+static struct rb_entry *__lookup_rb_tree_fast(struct rb_entry *cached_re,
+							unsigned int ofs)
+{
+	if (cached_re) {
+		if (cached_re->ofs <= ofs &&
+				cached_re->ofs + cached_re->len > ofs) {
+			return cached_re;
+		}
+	}
+	return NULL;
+}
+
+static struct rb_entry *__lookup_rb_tree_slow(struct rb_root *root,
+							unsigned int ofs)
+{
+	struct rb_node *node = root->rb_node;
+	struct rb_entry *re;
+
+	while (node) {
+		re = rb_entry(node, struct rb_entry, rb_node);
+
+		if (ofs < re->ofs)
+			node = node->rb_left;
+		else if (ofs >= re->ofs + re->len)
+			node = node->rb_right;
+		else
+			return re;
+	}
+	return NULL;
+}
+
+struct rb_entry *__lookup_rb_tree(struct rb_root *root,
+				struct rb_entry *cached_re, unsigned int ofs)
+{
+	struct rb_entry *re;
+
+	re = __lookup_rb_tree_fast(cached_re, ofs);
+	if (!re)
+		return __lookup_rb_tree_slow(root, ofs);
+
+	return re;
+}
+
+struct rb_node **__lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi,
+				struct rb_root *root, struct rb_node **parent,
+				unsigned int ofs)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_entry *re;
+
+	while (*p) {
+		*parent = *p;
+		re = rb_entry(*parent, struct rb_entry, rb_node);
+
+		if (ofs < re->ofs)
+			p = &(*p)->rb_left;
+		else if (ofs >= re->ofs + re->len)
+			p = &(*p)->rb_right;
+		else
+			f2fs_bug_on(sbi, 1);
+	}
+
+	return p;
+}
+
+/*
+ * lookup rb entry in position of @ofs in rb-tree,
+ * if hit, return the entry, otherwise, return NULL
+ * @prev_ex: extent before ofs
+ * @next_ex: extent after ofs
+ * @insert_p: insert point for new extent at ofs
+ * in order to simpfy the insertion after.
+ * tree must stay unchanged between lookup and insertion.
+ */
+struct rb_entry *__lookup_rb_tree_ret(struct rb_root *root,
+				struct rb_entry *cached_re,
+				unsigned int ofs,
+				struct rb_entry **prev_entry,
+				struct rb_entry **next_entry,
+				struct rb_node ***insert_p,
+				struct rb_node **insert_parent,
+				bool force)
+{
+	struct rb_node **pnode = &root->rb_node;
+	struct rb_node *parent = NULL, *tmp_node;
+	struct rb_entry *re = cached_re;
+
+	*insert_p = NULL;
+	*insert_parent = NULL;
+	*prev_entry = NULL;
+	*next_entry = NULL;
+
+	if (RB_EMPTY_ROOT(root))
+		return NULL;
+
+	if (re) {
+		if (re->ofs <= ofs && re->ofs + re->len > ofs)
+			goto lookup_neighbors;
+	}
+
+	while (*pnode) {
+		parent = *pnode;
+		re = rb_entry(*pnode, struct rb_entry, rb_node);
+
+		if (ofs < re->ofs)
+			pnode = &(*pnode)->rb_left;
+		else if (ofs >= re->ofs + re->len)
+			pnode = &(*pnode)->rb_right;
+		else
+			goto lookup_neighbors;
+	}
+
+	*insert_p = pnode;
+	*insert_parent = parent;
+
+	re = rb_entry(parent, struct rb_entry, rb_node);
+	tmp_node = parent;
+	if (parent && ofs > re->ofs)
+		tmp_node = rb_next(parent);
+	*next_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+
+	tmp_node = parent;
+	if (parent && ofs < re->ofs)
+		tmp_node = rb_prev(parent);
+	*prev_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+	return NULL;
+
+lookup_neighbors:
+	if (ofs == re->ofs || force) {
+		/* lookup prev node for merging backward later */
+		tmp_node = rb_prev(&re->rb_node);
+		*prev_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+	}
+	if (ofs == re->ofs + re->len - 1 || force) {
+		/* lookup next node for merging frontward later */
+		tmp_node = rb_next(&re->rb_node);
+		*next_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+	}
+	return re;
+}
+
+bool __check_rb_tree_consistence(struct f2fs_sb_info *sbi,
+						struct rb_root *root)
+{
+#ifdef CONFIG_F2FS_CHECK_FS
+	struct rb_node *cur = rb_first(root), *next;
+	struct rb_entry *cur_re, *next_re;
+
+	if (!cur)
+		return true;
+
+	while (cur) {
+		next = rb_next(cur);
+		if (!next)
+			return true;
+
+		cur_re = rb_entry(cur, struct rb_entry, rb_node);
+		next_re = rb_entry(next, struct rb_entry, rb_node);
+
+		if (cur_re->ofs + cur_re->len > next_re->ofs) {
+			f2fs_msg(sbi->sb, KERN_INFO, "inconsistent rbtree, "
+				"cur(%u, %u) next(%u, %u)",
+				cur_re->ofs, cur_re->len,
+				next_re->ofs, next_re->len);
+			return false;
+		}
+
+		cur = next;
+	}
+#endif
+	return true;
+}
+
 static struct kmem_cache *extent_tree_slab;
 static struct kmem_cache *extent_node_slab;
 
@@ -102,36 +275,6 @@ static struct extent_tree *__grab_extent_tree(struct inode *inode)
 	return et;
 }
 
-static struct extent_node *__lookup_extent_tree(struct f2fs_sb_info *sbi,
-				struct extent_tree *et, unsigned int fofs)
-{
-	struct rb_node *node = et->root.rb_node;
-	struct extent_node *en = et->cached_en;
-
-	if (en) {
-		struct extent_info *cei = &en->ei;
-
-		if (cei->fofs <= fofs && cei->fofs + cei->len > fofs) {
-			stat_inc_cached_node_hit(sbi);
-			return en;
-		}
-	}
-
-	while (node) {
-		en = rb_entry(node, struct extent_node, rb_node);
-
-		if (fofs < en->ei.fofs) {
-			node = node->rb_left;
-		} else if (fofs >= en->ei.fofs + en->ei.len) {
-			node = node->rb_right;
-		} else {
-			stat_inc_rbtree_node_hit(sbi);
-			return en;
-		}
-	}
-	return NULL;
-}
-
 static struct extent_node *__init_extent_tree(struct f2fs_sb_info *sbi,
 				struct extent_tree *et, struct extent_info *ei)
 {
@@ -237,8 +380,16 @@ static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
 		goto out;
 	}
 
-	en = __lookup_extent_tree(sbi, et, pgofs);
-	if (en) {
+	en = (struct extent_node *)__lookup_rb_tree(&et->root,
+				(struct rb_entry *)et->cached_en, pgofs);
+	if (!en)
+		goto out;
+
+	if (en == et->cached_en)
+		stat_inc_cached_node_hit(sbi);
+	else
+		stat_inc_rbtree_node_hit(sbi);
+
 	*ei = en->ei;
 	spin_lock(&sbi->extent_lock);
 	if (!list_empty(&en->list)) {
@@ -247,7 +398,6 @@ static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
 	}
 	spin_unlock(&sbi->extent_lock);
 	ret = true;
-	}
 out:
 	stat_inc_total_hit(sbi);
 	read_unlock(&et->lock);
@@ -256,83 +406,6 @@ static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
 	return ret;
 }
 
-
-/*
- * lookup extent at @fofs, if hit, return the extent
- * if not, return NULL and
- * @prev_ex: extent before fofs
- * @next_ex: extent after fofs
- * @insert_p: insert point for new extent at fofs
- * in order to simpfy the insertion after.
- * tree must stay unchanged between lookup and insertion.
- */
-static struct extent_node *__lookup_extent_tree_ret(struct extent_tree *et,
-				unsigned int fofs,
-				struct extent_node **prev_ex,
-				struct extent_node **next_ex,
-				struct rb_node ***insert_p,
-				struct rb_node **insert_parent)
-{
-	struct rb_node **pnode = &et->root.rb_node;
-	struct rb_node *parent = NULL, *tmp_node;
-	struct extent_node *en = et->cached_en;
-
-	*insert_p = NULL;
-	*insert_parent = NULL;
-	*prev_ex = NULL;
-	*next_ex = NULL;
-
-	if (RB_EMPTY_ROOT(&et->root))
-		return NULL;
-
-	if (en) {
-		struct extent_info *cei = &en->ei;
-
-		if (cei->fofs <= fofs && cei->fofs + cei->len > fofs)
-			goto lookup_neighbors;
-	}
-
-	while (*pnode) {
-		parent = *pnode;
-		en = rb_entry(*pnode, struct extent_node, rb_node);
-
-		if (fofs < en->ei.fofs)
-			pnode = &(*pnode)->rb_left;
-		else if (fofs >= en->ei.fofs + en->ei.len)
-			pnode = &(*pnode)->rb_right;
-		else
-			goto lookup_neighbors;
-	}
-
-	*insert_p = pnode;
-	*insert_parent = parent;
-
-	en = rb_entry(parent, struct extent_node, rb_node);
-	tmp_node = parent;
-	if (parent && fofs > en->ei.fofs)
-		tmp_node = rb_next(parent);
-	*next_ex = rb_entry_safe(tmp_node, struct extent_node, rb_node);
-
-	tmp_node = parent;
-	if (parent && fofs < en->ei.fofs)
-		tmp_node = rb_prev(parent);
-	*prev_ex = rb_entry_safe(tmp_node, struct extent_node, rb_node);
-	return NULL;
-
-lookup_neighbors:
-	if (fofs == en->ei.fofs) {
-		/* lookup prev node for merging backward later */
-		tmp_node = rb_prev(&en->rb_node);
-		*prev_ex = rb_entry_safe(tmp_node, struct extent_node, rb_node);
-	}
-	if (fofs == en->ei.fofs + en->ei.len - 1) {
-		/* lookup next node for merging frontward later */
-		tmp_node = rb_next(&en->rb_node);
-		*next_ex = rb_entry_safe(tmp_node, struct extent_node, rb_node);
-	}
-	return en;
-}
-
 static struct extent_node *__try_merge_extent_node(struct inode *inode,
 				struct extent_tree *et, struct extent_info *ei,
 				struct extent_node *prev_ex,
@@ -387,17 +460,7 @@ static struct extent_node *__insert_extent_tree(struct inode *inode,
 		goto do_insert;
 	}
 
-	while (*p) {
-		parent = *p;
-		en = rb_entry(parent, struct extent_node, rb_node);
-
-		if (ei->fofs < en->ei.fofs)
-			p = &(*p)->rb_left;
-		else if (ei->fofs >= en->ei.fofs + en->ei.len)
-			p = &(*p)->rb_right;
-		else
-			f2fs_bug_on(sbi, 1);
-	}
+	p = __lookup_rb_tree_for_insert(sbi, &et->root, &parent, ei->fofs);
 do_insert:
 	en = __attach_extent_node(sbi, et, ei, parent, p);
 	if (!en)
@@ -447,8 +510,11 @@ static void f2fs_update_extent_tree_range(struct inode *inode,
 	__drop_largest_extent(inode, fofs, len);
 
 	/* 1. lookup first extent node in range [fofs, fofs + len - 1] */
-	en = __lookup_extent_tree_ret(et, fofs, &prev_en, &next_en,
-					&insert_p, &insert_parent);
+	en = (struct extent_node *)__lookup_rb_tree_ret(&et->root,
+					(struct rb_entry *)et->cached_en, fofs,
+					(struct rb_entry **)&prev_en,
+					(struct rb_entry **)&next_en,
+					&insert_p, &insert_parent, false);
 	if (!en)
 		en = next_en;
 

fs/f2fs/gc.c

@@ -84,7 +84,7 @@ static int gc_thread_func(void *data)
 		stat_inc_bggc_count(sbi);
 
 		/* if return value is not zero, no victim was selected */
-		if (f2fs_gc(sbi, test_opt(sbi, FORCE_FG_GC), true))
+		if (f2fs_gc(sbi, test_opt(sbi, FORCE_FG_GC), true, NULL_SEGNO))
 			wait_ms = gc_th->no_gc_sleep_time;
 
 		trace_f2fs_background_gc(sbi->sb, wait_ms,
@@ -172,7 +172,11 @@ static void select_policy(struct f2fs_sb_info *sbi, int gc_type,
 	if (gc_type != FG_GC && p->max_search > sbi->max_victim_search)
 		p->max_search = sbi->max_victim_search;
 
-	p->offset = sbi->last_victim[p->gc_mode];
+	/* let's select beginning hot/small space first */
+	if (type == CURSEG_HOT_DATA || IS_NODESEG(type))
+		p->offset = 0;
+	else
+		p->offset = SIT_I(sbi)->last_victim[p->gc_mode];
 }
 
 static unsigned int get_max_cost(struct f2fs_sb_info *sbi,
@@ -182,7 +186,7 @@ static unsigned int get_max_cost(struct f2fs_sb_info *sbi,
 	if (p->alloc_mode == SSR)
 		return sbi->blocks_per_seg;
 	if (p->gc_mode == GC_GREEDY)
-		return sbi->blocks_per_seg * p->ofs_unit;
+		return 2 * sbi->blocks_per_seg * p->ofs_unit;
 	else if (p->gc_mode == GC_CB)
 		return UINT_MAX;
 	else /* No other gc_mode */
@@ -207,7 +211,7 @@ static unsigned int check_bg_victims(struct f2fs_sb_info *sbi)
 			continue;
 
 		clear_bit(secno, dirty_i->victim_secmap);
-		return secno * sbi->segs_per_sec;
+		return GET_SEG_FROM_SEC(sbi, secno);
 	}
 	return NULL_SEGNO;
 }
@@ -215,8 +219,8 @@ static unsigned int check_bg_victims(struct f2fs_sb_info *sbi)
 static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	unsigned int secno = GET_SECNO(sbi, segno);
-	unsigned int start = secno * sbi->segs_per_sec;
+	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
+	unsigned int start = GET_SEG_FROM_SEC(sbi, secno);
 	unsigned long long mtime = 0;
 	unsigned int vblocks;
 	unsigned char age = 0;
@@ -225,7 +229,7 @@ static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 
 	for (i = 0; i < sbi->segs_per_sec; i++)
 		mtime += get_seg_entry(sbi, start + i)->mtime;
-	vblocks = get_valid_blocks(sbi, segno, sbi->segs_per_sec);
+	vblocks = get_valid_blocks(sbi, segno, true);
 
 	mtime = div_u64(mtime, sbi->segs_per_sec);
 	vblocks = div_u64(vblocks, sbi->segs_per_sec);
@@ -248,7 +252,7 @@ static unsigned int get_greedy_cost(struct f2fs_sb_info *sbi,
 						unsigned int segno)
 {
 	unsigned int valid_blocks =
-			get_valid_blocks(sbi, segno, sbi->segs_per_sec);
+			get_valid_blocks(sbi, segno, true);
 
 	return IS_DATASEG(get_seg_entry(sbi, segno)->type) ?
 				valid_blocks * 2 : valid_blocks;
@@ -291,6 +295,7 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 		unsigned int *result, int gc_type, int type, char alloc_mode)
 {
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
+	struct sit_info *sm = SIT_I(sbi);
 	struct victim_sel_policy p;
 	unsigned int secno, last_victim;
 	unsigned int last_segment = MAIN_SEGS(sbi);
@@ -304,10 +309,18 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 	p.min_segno = NULL_SEGNO;
 	p.min_cost = get_max_cost(sbi, &p);
 
+	if (*result != NULL_SEGNO) {
+		if (IS_DATASEG(get_seg_entry(sbi, *result)->type) &&
+			get_valid_blocks(sbi, *result, false) &&
+			!sec_usage_check(sbi, GET_SEC_FROM_SEG(sbi, *result)))
+			p.min_segno = *result;
+		goto out;
+	}
+
 	if (p.max_search == 0)
 		goto out;
 
-	last_victim = sbi->last_victim[p.gc_mode];
+	last_victim = sm->last_victim[p.gc_mode];
 	if (p.alloc_mode == LFS && gc_type == FG_GC) {
 		p.min_segno = check_bg_victims(sbi);
 		if (p.min_segno != NULL_SEGNO)
@@ -320,9 +333,10 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 
 		segno = find_next_bit(p.dirty_segmap, last_segment, p.offset);
 		if (segno >= last_segment) {
-			if (sbi->last_victim[p.gc_mode]) {
-				last_segment = sbi->last_victim[p.gc_mode];
-				sbi->last_victim[p.gc_mode] = 0;
+			if (sm->last_victim[p.gc_mode]) {
+				last_segment =
+					sm->last_victim[p.gc_mode];
+				sm->last_victim[p.gc_mode] = 0;
 				p.offset = 0;
 				continue;
 			}
@@ -339,7 +353,7 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 			nsearched++;
 		}
 
-		secno = GET_SECNO(sbi, segno);
+		secno = GET_SEC_FROM_SEG(sbi, segno);
 
 		if (sec_usage_check(sbi, secno))
 			goto next;
@@ -357,17 +371,18 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 		}
 next:
 		if (nsearched >= p.max_search) {
-			if (!sbi->last_victim[p.gc_mode] && segno <= last_victim)
-				sbi->last_victim[p.gc_mode] = last_victim + 1;
+			if (!sm->last_victim[p.gc_mode] && segno <= last_victim)
+				sm->last_victim[p.gc_mode] = last_victim + 1;
 			else
-				sbi->last_victim[p.gc_mode] = segno + 1;
+				sm->last_victim[p.gc_mode] = segno + 1;
+			sm->last_victim[p.gc_mode] %= MAIN_SEGS(sbi);
 			break;
 		}
 	}
 	if (p.min_segno != NULL_SEGNO) {
 got_it:
 		if (p.alloc_mode == LFS) {
-			secno = GET_SECNO(sbi, p.min_segno);
+			secno = GET_SEC_FROM_SEG(sbi, p.min_segno);
 			if (gc_type == FG_GC)
 				sbi->cur_victim_sec = secno;
 			else
@@ -550,8 +565,10 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
 	get_node_info(sbi, nid, dni);
 
 	if (sum->version != dni->version) {
-		f2fs_put_page(node_page, 1);
-		return false;
+		f2fs_msg(sbi->sb, KERN_WARNING,
+				"%s: valid data with mismatched node version.",
+				__func__);
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
 	}
 
 	*nofs = ofs_of_node(node_page);
@@ -697,8 +714,10 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type,
 			.type = DATA,
 			.op = REQ_OP_WRITE,
 			.op_flags = REQ_SYNC,
+			.old_blkaddr = NULL_ADDR,
 			.page = page,
 			.encrypted_page = NULL,
+			.need_lock = true,
 		};
 		bool is_dirty = PageDirty(page);
 		int err;
@@ -890,7 +909,7 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
 					GET_SUM_BLOCK(sbi, segno));
 		f2fs_put_page(sum_page, 0);
 
-		if (get_valid_blocks(sbi, segno, 1) == 0 ||
+		if (get_valid_blocks(sbi, segno, false) == 0 ||
 				!PageUptodate(sum_page) ||
 				unlikely(f2fs_cp_error(sbi)))
 			goto next;
@@ -905,7 +924,6 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
 		 *   - mutex_lock(sentry_lock)     - change_curseg()
 		 *                                  - lock_page(sum_page)
 		 */
-
 		if (type == SUM_TYPE_NODE)
 			gc_node_segment(sbi, sum->entries, segno, gc_type);
 		else
@@ -924,7 +942,7 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
 	blk_finish_plug(&plug);
 
 	if (gc_type == FG_GC &&
-		get_valid_blocks(sbi, start_segno, sbi->segs_per_sec) == 0)
+		get_valid_blocks(sbi, start_segno, true) == 0)
 		sec_freed = 1;
 
 	stat_inc_call_count(sbi->stat_info);
@@ -932,13 +950,14 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi,
 	return sec_freed;
 }
 
-int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, bool background)
+int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
+			bool background, unsigned int segno)
 {
-	unsigned int segno;
 	int gc_type = sync ? FG_GC : BG_GC;
 	int sec_freed = 0;
 	int ret = -EINVAL;
 	struct cp_control cpc;
+	unsigned int init_segno = segno;
 	struct gc_inode_list gc_list = {
 		.ilist = LIST_HEAD_INIT(gc_list.ilist),
 		.iroot = RADIX_TREE_INIT(GFP_NOFS),
@@ -959,9 +978,11 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, bool background)
 		 * threshold, we can make them free by checkpoint. Then, we
 		 * secure free segments which doesn't need fggc any more.
 		 */
+		if (prefree_segments(sbi)) {
 			ret = write_checkpoint(sbi, &cpc);
 			if (ret)
 				goto stop;
+		}
 		if (has_not_enough_free_secs(sbi, 0, 0))
 			gc_type = FG_GC;
 	}
@@ -981,13 +1002,17 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, bool background)
 		sbi->cur_victim_sec = NULL_SEGNO;
 
 	if (!sync) {
-		if (has_not_enough_free_secs(sbi, sec_freed, 0))
+		if (has_not_enough_free_secs(sbi, sec_freed, 0)) {
+			segno = NULL_SEGNO;
 			goto gc_more;
+		}
 
 		if (gc_type == FG_GC)
 			ret = write_checkpoint(sbi, &cpc);
 	}
 stop:
+	SIT_I(sbi)->last_victim[ALLOC_NEXT] = 0;
+	SIT_I(sbi)->last_victim[FLUSH_DEVICE] = init_segno;
 	mutex_unlock(&sbi->gc_mutex);
 
 	put_gc_inode(&gc_list);
@@ -999,7 +1024,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, bool background)
 
 void build_gc_manager(struct f2fs_sb_info *sbi)
 {
-	u64 main_count, resv_count, ovp_count, blocks_per_sec;
+	u64 main_count, resv_count, ovp_count;
 
 	DIRTY_I(sbi)->v_ops = &default_v_ops;
 
@@ -1007,8 +1032,12 @@ void build_gc_manager(struct f2fs_sb_info *sbi)
 	main_count = SM_I(sbi)->main_segments << sbi->log_blocks_per_seg;
 	resv_count = SM_I(sbi)->reserved_segments << sbi->log_blocks_per_seg;
 	ovp_count = SM_I(sbi)->ovp_segments << sbi->log_blocks_per_seg;
-	blocks_per_sec = sbi->blocks_per_seg * sbi->segs_per_sec;
 
-	sbi->fggc_threshold = div64_u64((main_count - ovp_count) * blocks_per_sec,
-					(main_count - resv_count));
+	sbi->fggc_threshold = div64_u64((main_count - ovp_count) *
+				BLKS_PER_SEC(sbi), (main_count - resv_count));
+
+	/* give warm/cold data area from slower device */
+	if (sbi->s_ndevs && sbi->segs_per_sec == 1)
+		SIT_I(sbi)->last_victim[ALLOC_NEXT] =
+				GET_SEGNO(sbi, FDEV(0).end_blk) + 1;
 }

fs/f2fs/inline.c

@@ -63,19 +63,21 @@ void read_inline_data(struct page *page, struct page *ipage)
 		SetPageUptodate(page);
 }
 
-bool truncate_inline_inode(struct page *ipage, u64 from)
+void truncate_inline_inode(struct inode *inode, struct page *ipage, u64 from)
 {
 	void *addr;
 
 	if (from >= MAX_INLINE_DATA)
-		return false;
+		return;
 
 	addr = inline_data_addr(ipage);
 
 	f2fs_wait_on_page_writeback(ipage, NODE, true);
 	memset(addr + from, 0, MAX_INLINE_DATA - from);
 	set_page_dirty(ipage);
-	return true;
+
+	if (from == 0)
+		clear_inode_flag(inode, FI_DATA_EXIST);
 }
 
 int f2fs_read_inline_data(struct inode *inode, struct page *page)
@@ -135,6 +137,7 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page)
 	/* write data page to try to make data consistent */
 	set_page_writeback(page);
 	fio.old_blkaddr = dn->data_blkaddr;
+	set_inode_flag(dn->inode, FI_HOT_DATA);
 	write_data_page(dn, &fio);
 	f2fs_wait_on_page_writeback(page, DATA, true);
 	if (dirty) {
@@ -146,11 +149,11 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page)
 	set_inode_flag(dn->inode, FI_APPEND_WRITE);
 
 	/* clear inline data and flag after data writeback */
-	truncate_inline_inode(dn->inode_page, 0);
+	truncate_inline_inode(dn->inode, dn->inode_page, 0);
 	clear_inline_node(dn->inode_page);
 clear_out:
 	stat_dec_inline_inode(dn->inode);
-	f2fs_clear_inline_inode(dn->inode);
+	clear_inode_flag(dn->inode, FI_INLINE_DATA);
 	f2fs_put_dnode(dn);
 	return 0;
 }
@@ -267,9 +270,8 @@ bool recover_inline_data(struct inode *inode, struct page *npage)
 	if (f2fs_has_inline_data(inode)) {
 		ipage = get_node_page(sbi, inode->i_ino);
 		f2fs_bug_on(sbi, IS_ERR(ipage));
-		if (!truncate_inline_inode(ipage, 0))
-			return false;
-		f2fs_clear_inline_inode(inode);
+		truncate_inline_inode(inode, ipage, 0);
+		clear_inode_flag(inode, FI_INLINE_DATA);
 		f2fs_put_page(ipage, 1);
 	} else if (ri && (ri->i_inline & F2FS_INLINE_DATA)) {
 		if (truncate_blocks(inode, 0, false))
@@ -300,7 +302,7 @@ struct f2fs_dir_entry *find_in_inline_dir(struct inode *dir,
 
 	inline_dentry = inline_data_addr(ipage);
 
-	make_dentry_ptr(NULL, &d, (void *)inline_dentry, 2);
+	make_dentry_ptr_inline(NULL, &d, inline_dentry);
 	de = find_target_dentry(fname, namehash, NULL, &d);
 	unlock_page(ipage);
 	if (de)
@@ -319,7 +321,7 @@ int make_empty_inline_dir(struct inode *inode, struct inode *parent,
 
 	dentry_blk = inline_data_addr(ipage);
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 2);
+	make_dentry_ptr_inline(NULL, &d, dentry_blk);
 	do_make_empty_dir(inode, parent, &d);
 
 	set_page_dirty(ipage);
@@ -380,7 +382,7 @@ static int f2fs_move_inline_dirents(struct inode *dir, struct page *ipage,
 	set_page_dirty(page);
 
 	/* clear inline dir and flag after data writeback */
-	truncate_inline_inode(ipage, 0);
+	truncate_inline_inode(dir, ipage, 0);
 
 	stat_dec_inline_dir(dir);
 	clear_inode_flag(dir, FI_INLINE_DENTRY);
@@ -400,7 +402,7 @@ static int f2fs_add_inline_entries(struct inode *dir,
 	unsigned long bit_pos = 0;
 	int err = 0;
 
-	make_dentry_ptr(NULL, &d, (void *)inline_dentry, 2);
+	make_dentry_ptr_inline(NULL, &d, inline_dentry);
 
 	while (bit_pos < d.max) {
 		struct f2fs_dir_entry *de;
@@ -455,7 +457,7 @@ static int f2fs_move_rehashed_dirents(struct inode *dir, struct page *ipage,
 	}
 
 	memcpy(backup_dentry, inline_dentry, MAX_INLINE_DATA);
-	truncate_inline_inode(ipage, 0);
+	truncate_inline_inode(dir, ipage, 0);
 
 	unlock_page(ipage);
 
@@ -527,14 +529,12 @@ int f2fs_add_inline_entry(struct inode *dir, const struct qstr *new_name,
 			err = PTR_ERR(page);
 			goto fail;
 		}
-		if (f2fs_encrypted_inode(dir))
-			file_set_enc_name(inode);
 	}
 
 	f2fs_wait_on_page_writeback(ipage, NODE, true);
 
 	name_hash = f2fs_dentry_hash(new_name, NULL);
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 2);
+	make_dentry_ptr_inline(NULL, &d, dentry_blk);
 	f2fs_update_dentry(ino, mode, &d, new_name, name_hash, bit_pos);
 
 	set_page_dirty(ipage);
@@ -623,7 +623,7 @@ int f2fs_read_inline_dir(struct file *file, struct dir_context *ctx,
 
 	inline_dentry = inline_data_addr(ipage);
 
-	make_dentry_ptr(inode, &d, (void *)inline_dentry, 2);
+	make_dentry_ptr_inline(inode, &d, inline_dentry);
 
 	err = f2fs_fill_dentries(ctx, &d, 0, fstr);
 	if (!err)

fs/f2fs/inode.c

@@ -316,7 +316,6 @@ int update_inode_page(struct inode *inode)
 		} else if (err != -ENOENT) {
 			f2fs_stop_checkpoint(sbi, false);
 		}
-		f2fs_inode_synced(inode);
 		return 0;
 	}
 	ret = update_inode(inode, node_page);
@@ -339,7 +338,8 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 	 * We need to balance fs here to prevent from producing dirty node pages
 	 * during the urgent cleaning time when runing out of free sections.
 	 */
-	if (update_inode_page(inode) && wbc && wbc->nr_to_write)
+	update_inode_page(inode);
+	if (wbc && wbc->nr_to_write)
 		f2fs_balance_fs(sbi, true);
 	return 0;
 }
@@ -372,13 +372,6 @@ void f2fs_evict_inode(struct inode *inode)
 	if (inode->i_nlink || is_bad_inode(inode))
 		goto no_delete;
 
-#ifdef CONFIG_F2FS_FAULT_INJECTION
-	if (time_to_inject(sbi, FAULT_EVICT_INODE)) {
-		f2fs_show_injection_info(FAULT_EVICT_INODE);
-		goto no_delete;
-	}
-#endif
-
 	remove_ino_entry(sbi, inode->i_ino, APPEND_INO);
 	remove_ino_entry(sbi, inode->i_ino, UPDATE_INO);
 
@@ -389,6 +382,12 @@ void f2fs_evict_inode(struct inode *inode)
 	if (F2FS_HAS_BLOCKS(inode))
 		err = f2fs_truncate(inode);
 
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_EVICT_INODE)) {
+		f2fs_show_injection_info(FAULT_EVICT_INODE);
+		err = -EIO;
+	}
+#endif
 	if (!err) {
 		f2fs_lock_op(sbi);
 		err = remove_inode_page(inode);
@@ -411,7 +410,10 @@ void f2fs_evict_inode(struct inode *inode)
 	stat_dec_inline_dir(inode);
 	stat_dec_inline_inode(inode);
 
-	invalidate_mapping_pages(NODE_MAPPING(sbi), inode->i_ino, inode->i_ino);
+	/* ino == 0, if f2fs_new_inode() was failed t*/
+	if (inode->i_ino)
+		invalidate_mapping_pages(NODE_MAPPING(sbi), inode->i_ino,
+							inode->i_ino);
 	if (xnid)
 		invalidate_mapping_pages(NODE_MAPPING(sbi), xnid, xnid);
 	if (inode->i_nlink) {
@@ -448,6 +450,7 @@ void handle_failed_inode(struct inode *inode)
 	 * in a panic when flushing dirty inodes in gdirty_list.
 	 */
 	update_inode_page(inode);
+	f2fs_inode_synced(inode);
 
 	/* don't make bad inode, since it becomes a regular file. */
 	unlock_new_inode(inode);

fs/f2fs/namei.c

@@ -148,8 +148,6 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	inode->i_mapping->a_ops = &f2fs_dblock_aops;
 	ino = inode->i_ino;
 
-	f2fs_balance_fs(sbi, true);
-
 	f2fs_lock_op(sbi);
 	err = f2fs_add_link(dentry, inode);
 	if (err)
@@ -163,6 +161,8 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 
 	if (IS_DIRSYNC(dir))
 		f2fs_sync_fs(sbi->sb, 1);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 out:
 	handle_failed_inode(inode);
@@ -424,8 +424,6 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 	inode_nohighmem(inode);
 	inode->i_mapping->a_ops = &f2fs_dblock_aops;
 
-	f2fs_balance_fs(sbi, true);
-
 	f2fs_lock_op(sbi);
 	err = f2fs_add_link(dentry, inode);
 	if (err)
@@ -488,6 +486,8 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 	}
 
 	kfree(sd);
+
+	f2fs_balance_fs(sbi, true);
 	return err;
 out:
 	handle_failed_inode(inode);
@@ -509,8 +509,6 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	inode->i_mapping->a_ops = &f2fs_dblock_aops;
 	mapping_set_gfp_mask(inode->i_mapping, GFP_F2FS_HIGH_ZERO);
 
-	f2fs_balance_fs(sbi, true);
-
 	set_inode_flag(inode, FI_INC_LINK);
 	f2fs_lock_op(sbi);
 	err = f2fs_add_link(dentry, inode);
@@ -525,6 +523,8 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 
 	if (IS_DIRSYNC(dir))
 		f2fs_sync_fs(sbi->sb, 1);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 
 out_fail:
@@ -555,8 +555,6 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 	init_special_inode(inode, inode->i_mode, rdev);
 	inode->i_op = &f2fs_special_inode_operations;
 
-	f2fs_balance_fs(sbi, true);
-
 	f2fs_lock_op(sbi);
 	err = f2fs_add_link(dentry, inode);
 	if (err)
@@ -570,6 +568,8 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 
 	if (IS_DIRSYNC(dir))
 		f2fs_sync_fs(sbi->sb, 1);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 out:
 	handle_failed_inode(inode);
@@ -596,8 +596,6 @@ static int __f2fs_tmpfile(struct inode *dir, struct dentry *dentry,
 		inode->i_mapping->a_ops = &f2fs_dblock_aops;
 	}
 
-	f2fs_balance_fs(sbi, true);
-
 	f2fs_lock_op(sbi);
 	err = acquire_orphan_inode(sbi);
 	if (err)
@@ -623,6 +621,8 @@ static int __f2fs_tmpfile(struct inode *dir, struct dentry *dentry,
 	/* link_count was changed by d_tmpfile as well. */
 	f2fs_unlock_op(sbi);
 	unlock_new_inode(inode);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 
 release_out:
@@ -721,13 +721,6 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (err)
 			goto put_out_dir;
 
-		err = update_dent_inode(old_inode, new_inode,
-						&new_dentry->d_name);
-		if (err) {
-			release_orphan_inode(sbi);
-			goto put_out_dir;
-		}
-
 		f2fs_set_link(new_dir, new_entry, new_page, old_inode);
 
 		new_inode->i_ctime = current_time(new_inode);
@@ -780,8 +773,6 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	down_write(&F2FS_I(old_inode)->i_sem);
 	file_lost_pino(old_inode);
-	if (new_inode && file_enc_name(new_inode))
-		file_set_enc_name(old_inode);
 	up_write(&F2FS_I(old_inode)->i_sem);
 
 	old_inode->i_ctime = current_time(old_inode);
@@ -909,8 +900,8 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 		old_nlink = old_dir_entry ? -1 : 1;
 		new_nlink = -old_nlink;
 		err = -EMLINK;
-		if ((old_nlink > 0 && old_inode->i_nlink >= F2FS_LINK_MAX) ||
-			(new_nlink > 0 && new_inode->i_nlink >= F2FS_LINK_MAX))
+		if ((old_nlink > 0 && old_dir->i_nlink >= F2FS_LINK_MAX) ||
+			(new_nlink > 0 && new_dir->i_nlink >= F2FS_LINK_MAX))
 			goto out_new_dir;
 	}
 
@@ -918,18 +909,6 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	f2fs_lock_op(sbi);
 
-	err = update_dent_inode(old_inode, new_inode, &new_dentry->d_name);
-	if (err)
-		goto out_unlock;
-	if (file_enc_name(new_inode))
-		file_set_enc_name(old_inode);
-
-	err = update_dent_inode(new_inode, old_inode, &old_dentry->d_name);
-	if (err)
-		goto out_undo;
-	if (file_enc_name(old_inode))
-		file_set_enc_name(new_inode);
-
 	/* update ".." directory entry info of old dentry */
 	if (old_dir_entry)
 		f2fs_set_link(old_inode, old_dir_entry, old_dir_page, new_dir);
@@ -973,14 +952,6 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir))
 		f2fs_sync_fs(sbi->sb, 1);
 	return 0;
-out_undo:
-	/*
-	 * Still we may fail to recover name info of f2fs_inode here
-	 * Drop it, once its name is set as encrypted
-	 */
-	update_dent_inode(old_inode, old_inode, &old_dentry->d_name);
-out_unlock:
-	f2fs_unlock_op(sbi);
 out_new_dir:
 	if (new_dir_entry) {
 		f2fs_dentry_kunmap(new_inode, new_dir_page);

fs/f2fs/node.c

@@ -22,7 +22,7 @@
 #include "trace.h"
 #include <trace/events/f2fs.h>
 
-#define on_build_free_nids(nmi) mutex_is_locked(&nm_i->build_lock)
+#define on_build_free_nids(nmi) mutex_is_locked(&(nm_i)->build_lock)
 
 static struct kmem_cache *nat_entry_slab;
 static struct kmem_cache *free_nid_slab;
@@ -63,8 +63,9 @@ bool available_free_memory(struct f2fs_sb_info *sbi, int type)
 		int i;
 
 		for (i = 0; i <= UPDATE_INO; i++)
-			mem_size += (sbi->im[i].ino_num *
-				sizeof(struct ino_entry)) >> PAGE_SHIFT;
+			mem_size += sbi->im[i].ino_num *
+						sizeof(struct ino_entry);
+		mem_size >>= PAGE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1);
 	} else if (type == EXTENT_CACHE) {
 		mem_size = (atomic_read(&sbi->total_ext_tree) *
@@ -177,18 +178,12 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
 }
 
 static void __clear_nat_cache_dirty(struct f2fs_nm_info *nm_i,
-						struct nat_entry *ne)
+		struct nat_entry_set *set, struct nat_entry *ne)
 {
-	nid_t set = NAT_BLOCK_OFFSET(ne->ni.nid);
-	struct nat_entry_set *head;
-
-	head = radix_tree_lookup(&nm_i->nat_set_root, set);
-	if (head) {
 	list_move_tail(&ne->list, &nm_i->nat_entries);
 	set_nat_flag(ne, IS_DIRTY, false);
-		head->entry_cnt--;
+	set->entry_cnt--;
 	nm_i->dirty_nat_cnt--;
-	}
 }
 
 static unsigned int __gang_lookup_nat_set(struct f2fs_nm_info *nm_i,
@@ -381,6 +376,7 @@ void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
 	struct page *page = NULL;
 	struct f2fs_nat_entry ne;
 	struct nat_entry *e;
+	pgoff_t index;
 	int i;
 
 	ni->nid = nid;
@@ -406,17 +402,21 @@ void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
 		node_info_from_raw_nat(ni, &ne);
 	}
 	up_read(&curseg->journal_rwsem);
-	if (i >= 0)
+	if (i >= 0) {
+		up_read(&nm_i->nat_tree_lock);
 		goto cache;
+	}
 
 	/* Fill node_info from nat page */
-	page = get_current_nat_page(sbi, start_nid);
+	index = current_nat_addr(sbi, nid);
+	up_read(&nm_i->nat_tree_lock);
+
+	page = get_meta_page(sbi, index);
 	nat_blk = (struct f2fs_nat_block *)page_address(page);
 	ne = nat_blk->entries[nid - start_nid];
 	node_info_from_raw_nat(ni, &ne);
 	f2fs_put_page(page, 1);
 cache:
-	up_read(&nm_i->nat_tree_lock);
 	/* cache nat entry */
 	down_write(&nm_i->nat_tree_lock);
 	cache_nat_entry(sbi, nid, &ne);
@@ -1463,6 +1463,9 @@ int fsync_node_pages(struct f2fs_sb_info *sbi, struct inode *inode,
 			f2fs_wait_on_page_writeback(page, NODE, true);
 			BUG_ON(PageWriteback(page));
 
+			set_fsync_mark(page, 0);
+			set_dentry_mark(page, 0);
+
 			if (!atomic || page == last_page) {
 				set_fsync_mark(page, 1);
 				if (IS_INODE(page)) {
@@ -1766,40 +1769,67 @@ static void __remove_nid_from_list(struct f2fs_sb_info *sbi,
 static bool add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
-	struct free_nid *i;
+	struct free_nid *i, *e;
 	struct nat_entry *ne;
-	int err;
+	int err = -EINVAL;
+	bool ret = false;
 
 	/* 0 nid should not be used */
 	if (unlikely(nid == 0))
 		return false;
 
-	if (build) {
-		/* do not add allocated nids */
-		ne = __lookup_nat_cache(nm_i, nid);
-		if (ne && (!get_nat_flag(ne, IS_CHECKPOINTED) ||
-				nat_get_blkaddr(ne) != NULL_ADDR))
-			return false;
-	}
-
 	i = f2fs_kmem_cache_alloc(free_nid_slab, GFP_NOFS);
 	i->nid = nid;
 	i->state = NID_NEW;
 
-	if (radix_tree_preload(GFP_NOFS)) {
-		kmem_cache_free(free_nid_slab, i);
-		return true;
-	}
+	if (radix_tree_preload(GFP_NOFS))
+		goto err;
 
 	spin_lock(&nm_i->nid_list_lock);
+
+	if (build) {
+		/*
+		 *   Thread A             Thread B
+		 *  - f2fs_create
+		 *   - f2fs_new_inode
+		 *    - alloc_nid
+		 *     - __insert_nid_to_list(ALLOC_NID_LIST)
+		 *                     - f2fs_balance_fs_bg
+		 *                      - build_free_nids
+		 *                       - __build_free_nids
+		 *                        - scan_nat_page
+		 *                         - add_free_nid
+		 *                          - __lookup_nat_cache
+		 *  - f2fs_add_link
+		 *   - init_inode_metadata
+		 *    - new_inode_page
+		 *     - new_node_page
+		 *      - set_node_addr
+		 *  - alloc_nid_done
+		 *   - __remove_nid_from_list(ALLOC_NID_LIST)
+		 *                         - __insert_nid_to_list(FREE_NID_LIST)
+		 */
+		ne = __lookup_nat_cache(nm_i, nid);
+		if (ne && (!get_nat_flag(ne, IS_CHECKPOINTED) ||
+				nat_get_blkaddr(ne) != NULL_ADDR))
+			goto err_out;
+
+		e = __lookup_free_nid_list(nm_i, nid);
+		if (e) {
+			if (e->state == NID_NEW)
+				ret = true;
+			goto err_out;
+		}
+	}
+	ret = true;
 	err = __insert_nid_to_list(sbi, i, FREE_NID_LIST, true);
+err_out:
 	spin_unlock(&nm_i->nid_list_lock);
 	radix_tree_preload_end();
-	if (err) {
+err:
+	if (err)
 		kmem_cache_free(free_nid_slab, i);
-		return true;
-	}
-	return true;
+	return ret;
 }
 
 static void remove_free_nid(struct f2fs_sb_info *sbi, nid_t nid)
@@ -1821,7 +1851,7 @@ static void remove_free_nid(struct f2fs_sb_info *sbi, nid_t nid)
 }
 
 static void update_free_nid_bitmap(struct f2fs_sb_info *sbi, nid_t nid,
-			bool set, bool build, bool locked)
+							bool set, bool build)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	unsigned int nat_ofs = NAT_BLOCK_OFFSET(nid);
@@ -1835,14 +1865,10 @@ static void update_free_nid_bitmap(struct f2fs_sb_info *sbi, nid_t nid,
 	else
 		__clear_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]);
 
-	if (!locked)
-		spin_lock(&nm_i->free_nid_lock);
 	if (set)
 		nm_i->free_nid_count[nat_ofs]++;
 	else if (!build)
 		nm_i->free_nid_count[nat_ofs]--;
-	if (!locked)
-		spin_unlock(&nm_i->free_nid_lock);
 }
 
 static void scan_nat_page(struct f2fs_sb_info *sbi,
@@ -1871,7 +1897,9 @@ static void scan_nat_page(struct f2fs_sb_info *sbi,
 		f2fs_bug_on(sbi, blk_addr == NEW_ADDR);
 		if (blk_addr == NULL_ADDR)
 			freed = add_free_nid(sbi, start_nid, true);
-		update_free_nid_bitmap(sbi, start_nid, freed, true, false);
+		spin_lock(&NM_I(sbi)->nid_list_lock);
+		update_free_nid_bitmap(sbi, start_nid, freed, true);
+		spin_unlock(&NM_I(sbi)->nid_list_lock);
 	}
 }
 
@@ -1927,6 +1955,9 @@ static void __build_free_nids(struct f2fs_sb_info *sbi, bool sync, bool mount)
 	int i = 0;
 	nid_t nid = nm_i->next_scan_nid;
 
+	if (unlikely(nid >= nm_i->max_nid))
+		nid = 0;
+
 	/* Enough entries */
 	if (nm_i->nid_cnt[FREE_NID_LIST] >= NAT_ENTRY_PER_BLOCK)
 		return;
@@ -2026,7 +2057,7 @@ bool alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid)
 		__insert_nid_to_list(sbi, i, ALLOC_NID_LIST, false);
 		nm_i->available_nids--;
 
-		update_free_nid_bitmap(sbi, *nid, false, false, false);
+		update_free_nid_bitmap(sbi, *nid, false, false);
 
 		spin_unlock(&nm_i->nid_list_lock);
 		return true;
@@ -2082,7 +2113,7 @@ void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid)
 
 	nm_i->available_nids++;
 
-	update_free_nid_bitmap(sbi, nid, true, false, false);
+	update_free_nid_bitmap(sbi, nid, true, false);
 
 	spin_unlock(&nm_i->nid_list_lock);
 
@@ -2407,16 +2438,16 @@ static void __flush_nat_entry_set(struct f2fs_sb_info *sbi,
 		}
 		raw_nat_from_node_info(raw_ne, &ne->ni);
 		nat_reset_flag(ne);
-		__clear_nat_cache_dirty(NM_I(sbi), ne);
+		__clear_nat_cache_dirty(NM_I(sbi), set, ne);
 		if (nat_get_blkaddr(ne) == NULL_ADDR) {
 			add_free_nid(sbi, nid, false);
 			spin_lock(&NM_I(sbi)->nid_list_lock);
 			NM_I(sbi)->available_nids++;
-			update_free_nid_bitmap(sbi, nid, true, false, false);
+			update_free_nid_bitmap(sbi, nid, true, false);
 			spin_unlock(&NM_I(sbi)->nid_list_lock);
 		} else {
 			spin_lock(&NM_I(sbi)->nid_list_lock);
-			update_free_nid_bitmap(sbi, nid, false, false, false);
+			update_free_nid_bitmap(sbi, nid, false, false);
 			spin_unlock(&NM_I(sbi)->nid_list_lock);
 		}
 	}
@@ -2428,10 +2459,11 @@ static void __flush_nat_entry_set(struct f2fs_sb_info *sbi,
 		f2fs_put_page(page, 1);
 	}
 
-	f2fs_bug_on(sbi, set->entry_cnt);
-
+	/* Allow dirty nats by node block allocation in write_begin */
+	if (!set->entry_cnt) {
 		radix_tree_delete(&NM_I(sbi)->nat_set_root, set->set);
 		kmem_cache_free(nat_entry_set_slab, set);
+	}
 }
 
 /*
@@ -2476,8 +2508,7 @@ void flush_nat_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 		__flush_nat_entry_set(sbi, set, cpc);
 
 	up_write(&nm_i->nat_tree_lock);
-
-	f2fs_bug_on(sbi, nm_i->dirty_nat_cnt);
+	/* Allow dirty nats by node block allocation in write_begin */
 }
 
 static int __get_nat_bitmaps(struct f2fs_sb_info *sbi)
@@ -2541,10 +2572,10 @@ inline void load_free_nid_bitmap(struct f2fs_sb_info *sbi)
 		nid = i * NAT_ENTRY_PER_BLOCK;
 		last_nid = (i + 1) * NAT_ENTRY_PER_BLOCK;
 
-		spin_lock(&nm_i->free_nid_lock);
+		spin_lock(&NM_I(sbi)->nid_list_lock);
 		for (; nid < last_nid; nid++)
-			update_free_nid_bitmap(sbi, nid, true, true, true);
-		spin_unlock(&nm_i->free_nid_lock);
+			update_free_nid_bitmap(sbi, nid, true, true);
+		spin_unlock(&NM_I(sbi)->nid_list_lock);
 	}
 
 	for (i = 0; i < nm_i->nat_blocks; i++) {
@@ -2635,9 +2666,6 @@ static int init_free_nid_cache(struct f2fs_sb_info *sbi)
 					sizeof(unsigned short), GFP_KERNEL);
 	if (!nm_i->free_nid_count)
 		return -ENOMEM;
-
-	spin_lock_init(&nm_i->free_nid_lock);
-
 	return 0;
 }
 

fs/f2fs/node.h

@@ -9,10 +9,10 @@
  * published by the Free Software Foundation.
  */
 /* start node id of a node block dedicated to the given node id */
-#define	START_NID(nid) ((nid / NAT_ENTRY_PER_BLOCK) * NAT_ENTRY_PER_BLOCK)
+#define	START_NID(nid) (((nid) / NAT_ENTRY_PER_BLOCK) * NAT_ENTRY_PER_BLOCK)
 
 /* node block offset on the NAT area dedicated to the given start node id */
-#define	NAT_BLOCK_OFFSET(start_nid) (start_nid / NAT_ENTRY_PER_BLOCK)
+#define	NAT_BLOCK_OFFSET(start_nid) ((start_nid) / NAT_ENTRY_PER_BLOCK)
 
 /* # of pages to perform synchronous readahead before building free nids */
 #define FREE_NID_PAGES	8
@@ -62,16 +62,16 @@ struct nat_entry {
 	struct node_info ni;	/* in-memory node information */
 };
 
-#define nat_get_nid(nat)		(nat->ni.nid)
-#define nat_set_nid(nat, n)		(nat->ni.nid = n)
-#define nat_get_blkaddr(nat)		(nat->ni.blk_addr)
-#define nat_set_blkaddr(nat, b)		(nat->ni.blk_addr = b)
-#define nat_get_ino(nat)		(nat->ni.ino)
-#define nat_set_ino(nat, i)		(nat->ni.ino = i)
-#define nat_get_version(nat)		(nat->ni.version)
-#define nat_set_version(nat, v)		(nat->ni.version = v)
+#define nat_get_nid(nat)		((nat)->ni.nid)
+#define nat_set_nid(nat, n)		((nat)->ni.nid = (n))
+#define nat_get_blkaddr(nat)		((nat)->ni.blk_addr)
+#define nat_set_blkaddr(nat, b)		((nat)->ni.blk_addr = (b))
+#define nat_get_ino(nat)		((nat)->ni.ino)
+#define nat_set_ino(nat, i)		((nat)->ni.ino = (i))
+#define nat_get_version(nat)		((nat)->ni.version)
+#define nat_set_version(nat, v)		((nat)->ni.version = (v))
 
-#define inc_node_version(version)	(++version)
+#define inc_node_version(version)	(++(version))
 
 static inline void copy_node_info(struct node_info *dst,
 						struct node_info *src)
@@ -200,13 +200,16 @@ static inline pgoff_t current_nat_addr(struct f2fs_sb_info *sbi, nid_t start)
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	pgoff_t block_off;
 	pgoff_t block_addr;
-	int seg_off;
 
+	/*
+	 * block_off = segment_off * 512 + off_in_segment
+	 * OLD = (segment_off * 512) * 2 + off_in_segment
+	 * NEW = 2 * (segment_off * 512 + off_in_segment) - off_in_segment
+	 */
 	block_off = NAT_BLOCK_OFFSET(start);
-	seg_off = block_off >> sbi->log_blocks_per_seg;
 
 	block_addr = (pgoff_t)(nm_i->nat_blkaddr +
-		(seg_off << sbi->log_blocks_per_seg << 1) +
+		(block_off << 1) -
 		(block_off & (sbi->blocks_per_seg - 1)));
 
 	if (f2fs_test_bit(block_off, nm_i->nat_bitmap))

fs/f2fs/recovery.c

@@ -198,7 +198,8 @@ static void recover_inode(struct inode *inode, struct page *page)
 			ino_of_node(page), name);
 }
 
-static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head)
+static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
+				bool check_only)
 {
 	struct curseg_info *curseg;
 	struct page *page = NULL;
@@ -225,7 +226,8 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head)
 
 		entry = get_fsync_inode(head, ino_of_node(page));
 		if (!entry) {
-			if (IS_INODE(page) && is_dent_dnode(page)) {
+			if (!check_only &&
+					IS_INODE(page) && is_dent_dnode(page)) {
 				err = recover_inode_page(sbi, page);
 				if (err)
 					break;
@@ -569,7 +571,7 @@ int recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	mutex_lock(&sbi->cp_mutex);
 
 	/* step #1: find fsynced inode numbers */
-	err = find_fsync_dnodes(sbi, &inode_list);
+	err = find_fsync_dnodes(sbi, &inode_list, check_only);
 	if (err || list_empty(&inode_list))
 		goto out;
 

fs/f2fs/super.c

@@ -49,6 +49,7 @@ char *fault_name[FAULT_MAX] = {
 	[FAULT_BLOCK]		= "no more block",
 	[FAULT_DIR_DEPTH]	= "too big dir depth",
 	[FAULT_EVICT_INODE]	= "evict_inode fail",
+	[FAULT_TRUNCATE]	= "truncate fail",
 	[FAULT_IO]		= "IO error",
 	[FAULT_CHECKPOINT]	= "checkpoint error",
 };
@@ -82,6 +83,7 @@ enum {
 	Opt_discard,
 	Opt_nodiscard,
 	Opt_noheap,
+	Opt_heap,
 	Opt_user_xattr,
 	Opt_nouser_xattr,
 	Opt_acl,
@@ -116,6 +118,7 @@ static match_table_t f2fs_tokens = {
 	{Opt_discard, "discard"},
 	{Opt_nodiscard, "nodiscard"},
 	{Opt_noheap, "no_heap"},
+	{Opt_heap, "heap"},
 	{Opt_user_xattr, "user_xattr"},
 	{Opt_nouser_xattr, "nouser_xattr"},
 	{Opt_acl, "acl"},
@@ -293,6 +296,7 @@ F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, ipu_policy, ipu_policy);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_ipu_util, min_ipu_util);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_hot_blocks, min_hot_blocks);
 F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ram_thresh, ram_thresh);
 F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ra_nid_pages, ra_nid_pages);
 F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, dirty_nats_ratio, dirty_nats_ratio);
@@ -318,6 +322,7 @@ static struct attribute *f2fs_attrs[] = {
 	ATTR_LIST(ipu_policy),
 	ATTR_LIST(min_ipu_util),
 	ATTR_LIST(min_fsync_blocks),
+	ATTR_LIST(min_hot_blocks),
 	ATTR_LIST(max_victim_search),
 	ATTR_LIST(dir_level),
 	ATTR_LIST(ram_thresh),
@@ -436,6 +441,9 @@ static int parse_options(struct super_block *sb, char *options)
 		case Opt_noheap:
 			set_opt(sbi, NOHEAP);
 			break;
+		case Opt_heap:
+			clear_opt(sbi, NOHEAP);
+			break;
 #ifdef CONFIG_F2FS_FS_XATTR
 		case Opt_user_xattr:
 			set_opt(sbi, XATTR_USER);
@@ -787,7 +795,14 @@ static void f2fs_put_super(struct super_block *sb)
 	}
 
 	/* be sure to wait for any on-going discard commands */
-	f2fs_wait_discard_bio(sbi, NULL_ADDR);
+	f2fs_wait_discard_bios(sbi);
+
+	if (!sbi->discard_blks) {
+		struct cp_control cpc = {
+			.reason = CP_UMOUNT | CP_TRIMMED,
+		};
+		write_checkpoint(sbi, &cpc);
+	}
 
 	/* write_checkpoint can update stat informaion */
 	f2fs_destroy_stats(sbi);
@@ -913,7 +928,9 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 	if (test_opt(sbi, DISCARD))
 		seq_puts(seq, ",discard");
 	if (test_opt(sbi, NOHEAP))
-		seq_puts(seq, ",no_heap_alloc");
+		seq_puts(seq, ",no_heap");
+	else
+		seq_puts(seq, ",heap");
 #ifdef CONFIG_F2FS_FS_XATTR
 	if (test_opt(sbi, XATTR_USER))
 		seq_puts(seq, ",user_xattr");
@@ -986,7 +1003,7 @@ static int segment_info_seq_show(struct seq_file *seq, void *offset)
 		if ((i % 10) == 0)
 			seq_printf(seq, "%-10d", i);
 		seq_printf(seq, "%d|%-3u", se->type,
-					get_valid_blocks(sbi, i, 1));
+					get_valid_blocks(sbi, i, false));
 		if ((i % 10) == 9 || i == (total_segs - 1))
 			seq_putc(seq, '\n');
 		else
@@ -1012,7 +1029,7 @@ static int segment_bits_seq_show(struct seq_file *seq, void *offset)
 
 		seq_printf(seq, "%-10d", i);
 		seq_printf(seq, "%d|%-3u|", se->type,
-					get_valid_blocks(sbi, i, 1));
+					get_valid_blocks(sbi, i, false));
 		for (j = 0; j < SIT_VBLOCK_MAP_SIZE; j++)
 			seq_printf(seq, " %.2x", se->cur_valid_map[j]);
 		seq_putc(seq, '\n');
@@ -1046,6 +1063,7 @@ static void default_options(struct f2fs_sb_info *sbi)
 	set_opt(sbi, INLINE_DATA);
 	set_opt(sbi, INLINE_DENTRY);
 	set_opt(sbi, EXTENT_CACHE);
+	set_opt(sbi, NOHEAP);
 	sbi->sb->s_flags |= MS_LAZYTIME;
 	set_opt(sbi, FLUSH_MERGE);
 	if (f2fs_sb_mounted_blkzoned(sbi->sb)) {
@@ -1307,7 +1325,7 @@ static int __f2fs_commit_super(struct buffer_head *bh,
 	unlock_buffer(bh);
 
 	/* it's rare case, we can do fua all the time */
-	return __sync_dirty_buffer(bh, REQ_PREFLUSH | REQ_FUA);
+	return __sync_dirty_buffer(bh, REQ_SYNC | REQ_PREFLUSH | REQ_FUA);
 }
 
 static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi,
@@ -1483,6 +1501,13 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
 		return 1;
 	}
 
+	if (le32_to_cpu(raw_super->segment_count) > F2FS_MAX_SEGMENT) {
+		f2fs_msg(sb, KERN_INFO,
+			"Invalid segment count (%u)",
+			le32_to_cpu(raw_super->segment_count));
+		return 1;
+	}
+
 	/* check CP/SIT/NAT/SSA/MAIN_AREA area boundary */
 	if (sanity_check_area_boundary(sbi, bh))
 		return 1;
@@ -1555,6 +1580,8 @@ static void init_sb_info(struct f2fs_sb_info *sbi)
 	for (i = 0; i < NR_COUNT_TYPE; i++)
 		atomic_set(&sbi->nr_pages[i], 0);
 
+	atomic_set(&sbi->wb_sync_req, 0);
+
 	INIT_LIST_HEAD(&sbi->s_list);
 	mutex_init(&sbi->umount_mutex);
 	mutex_init(&sbi->wio_mutex[NODE]);
@@ -1917,6 +1944,7 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	mutex_init(&sbi->gc_mutex);
 	mutex_init(&sbi->cp_mutex);
 	init_rwsem(&sbi->node_write);
+	init_rwsem(&sbi->node_change);
 
 	/* disallow all the data/node/meta page writes */
 	set_sbi_flag(sbi, SBI_POR_DOING);
@@ -2022,6 +2050,10 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 
 	f2fs_join_shrinker(sbi);
 
+	err = f2fs_build_stats(sbi);
+	if (err)
+		goto free_nm;
+
 	/* if there are nt orphan nodes free them */
 	err = recover_orphan_inodes(sbi);
 	if (err)
@@ -2046,10 +2078,6 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 		goto free_root_inode;
 	}
 
-	err = f2fs_build_stats(sbi);
-	if (err)
-		goto free_root_inode;
-
 	if (f2fs_proc_root)
 		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);
 
@@ -2143,7 +2171,6 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 		remove_proc_entry("segment_bits", sbi->s_proc);
 		remove_proc_entry(sb->s_id, f2fs_proc_root);
 	}
-	f2fs_destroy_stats(sbi);
 free_root_inode:
 	dput(sb->s_root);
 	sb->s_root = NULL;
@@ -2161,6 +2188,7 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 	truncate_inode_pages_final(META_MAPPING(sbi));
 	iput(sbi->node_inode);
 	mutex_unlock(&sbi->umount_mutex);
+	f2fs_destroy_stats(sbi);
 free_nm:
 	destroy_node_manager(sbi);
 free_sm:

fs/f2fs/trace.c

@@ -59,7 +59,7 @@ void f2fs_trace_pid(struct page *page)
 	pid_t pid = task_pid_nr(current);
 	void *p;
 
-	page->private = pid;
+	set_page_private(page, (unsigned long)pid);
 
 	if (radix_tree_preload(GFP_NOFS))
 		return;
@@ -138,7 +138,7 @@ static unsigned int gang_lookup_pids(pid_t *results, unsigned long first_index,
 
 	radix_tree_for_each_slot(slot, &pids, &iter, first_index) {
 		results[ret] = iter.index;
-		if (++ret == PIDVEC_SIZE)
+		if (++ret == max_items)
 			break;
 	}
 	return ret;

fs/f2fs/xattr.c

@@ -250,15 +250,13 @@ static int lookup_all_xattrs(struct inode *inode, struct page *ipage,
 	void *cur_addr, *txattr_addr, *last_addr = NULL;
 	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
 	unsigned int size = xnid ? VALID_XATTR_BLOCK_SIZE : 0;
-	unsigned int inline_size = 0;
+	unsigned int inline_size = inline_xattr_size(inode);
 	int err = 0;
 
-	inline_size = inline_xattr_size(inode);
-
 	if (!size && !inline_size)
 		return -ENODATA;
 
-	txattr_addr = kzalloc(inline_size + size + sizeof(__u32),
+	txattr_addr = kzalloc(inline_size + size + XATTR_PADDING_SIZE,
 							GFP_F2FS_ZERO);
 	if (!txattr_addr)
 		return -ENOMEM;
@@ -328,13 +326,14 @@ static int read_all_xattrs(struct inode *inode, struct page *ipage,
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct f2fs_xattr_header *header;
-	size_t size = PAGE_SIZE, inline_size = 0;
+	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
+	unsigned int size = VALID_XATTR_BLOCK_SIZE;
+	unsigned int inline_size = inline_xattr_size(inode);
 	void *txattr_addr;
 	int err;
 
-	inline_size = inline_xattr_size(inode);
-
-	txattr_addr = kzalloc(inline_size + size, GFP_F2FS_ZERO);
+	txattr_addr = kzalloc(inline_size + size + XATTR_PADDING_SIZE,
+							GFP_F2FS_ZERO);
 	if (!txattr_addr)
 		return -ENOMEM;
 
@@ -358,19 +357,19 @@ static int read_all_xattrs(struct inode *inode, struct page *ipage,
 	}
 
 	/* read from xattr node block */
-	if (F2FS_I(inode)->i_xattr_nid) {
+	if (xnid) {
 		struct page *xpage;
 		void *xattr_addr;
 
 		/* The inode already has an extended attribute block. */
-		xpage = get_node_page(sbi, F2FS_I(inode)->i_xattr_nid);
+		xpage = get_node_page(sbi, xnid);
 		if (IS_ERR(xpage)) {
 			err = PTR_ERR(xpage);
 			goto fail;
 		}
 
 		xattr_addr = page_address(xpage);
-		memcpy(txattr_addr + inline_size, xattr_addr, PAGE_SIZE);
+		memcpy(txattr_addr + inline_size, xattr_addr, size);
 		f2fs_put_page(xpage, 1);
 	}
 
@@ -392,14 +391,12 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 				void *txattr_addr, struct page *ipage)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	size_t inline_size = 0;
+	size_t inline_size = inline_xattr_size(inode);
 	void *xattr_addr;
 	struct page *xpage;
 	nid_t new_nid = 0;
 	int err;
 
-	inline_size = inline_xattr_size(inode);
-
 	if (hsize > inline_size && !F2FS_I(inode)->i_xattr_nid)
 		if (!alloc_nid(sbi, &new_nid))
 			return -ENOSPC;
@@ -454,7 +451,7 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 	}
 
 	xattr_addr = page_address(xpage);
-	memcpy(xattr_addr, txattr_addr + inline_size, MAX_XATTR_BLOCK_SIZE);
+	memcpy(xattr_addr, txattr_addr + inline_size, VALID_XATTR_BLOCK_SIZE);
 	set_page_dirty(xpage);
 	f2fs_put_page(xpage, 1);
 
@@ -546,7 +543,9 @@ static bool f2fs_xattr_value_same(struct f2fs_xattr_entry *entry,
 					const void *value, size_t size)
 {
 	void *pval = entry->e_name + entry->e_name_len;
-	return (entry->e_value_size == size) && !memcmp(pval, value, size);
+
+	return (le16_to_cpu(entry->e_value_size) == size) &&
+					!memcmp(pval, value, size);
 }
 
 static int __f2fs_setxattr(struct inode *inode, int index,

fs/f2fs/xattr.h

@@ -58,10 +58,10 @@ struct f2fs_xattr_entry {
 #define XATTR_FIRST_ENTRY(ptr)	(XATTR_ENTRY(XATTR_HDR(ptr) + 1))
 #define XATTR_ROUND		(3)
 
-#define XATTR_ALIGN(size)	((size + XATTR_ROUND) & ~XATTR_ROUND)
+#define XATTR_ALIGN(size)	(((size) + XATTR_ROUND) & ~XATTR_ROUND)
 
 #define ENTRY_SIZE(entry) (XATTR_ALIGN(sizeof(struct f2fs_xattr_entry) + \
-			entry->e_name_len + le16_to_cpu(entry->e_value_size)))
+			(entry)->e_name_len + le16_to_cpu((entry)->e_value_size)))
 
 #define XATTR_NEXT_ENTRY(entry)	((struct f2fs_xattr_entry *)((char *)(entry) +\
 			ENTRY_SIZE(entry)))
@@ -72,8 +72,8 @@ struct f2fs_xattr_entry {
 		for (entry = XATTR_FIRST_ENTRY(addr);\
 				!IS_XATTR_LAST_ENTRY(entry);\
 				entry = XATTR_NEXT_ENTRY(entry))
-#define MAX_XATTR_BLOCK_SIZE	(PAGE_SIZE - sizeof(struct node_footer))
-#define VALID_XATTR_BLOCK_SIZE	(MAX_XATTR_BLOCK_SIZE - sizeof(__u32))
+#define VALID_XATTR_BLOCK_SIZE	(PAGE_SIZE - sizeof(struct node_footer))
+#define XATTR_PADDING_SIZE	(sizeof(__u32))
 #define MIN_OFFSET(i)		XATTR_ALIGN(inline_xattr_size(i) +	\
 						VALID_XATTR_BLOCK_SIZE)
 

include/linux/f2fs_fs.h

@@ -32,9 +32,9 @@
 /* 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)
+#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)
 
 #define F2FS_IO_SIZE(sbi)	(1 << (sbi)->write_io_size_bits) /* Blocks */
 #define F2FS_IO_SIZE_KB(sbi)	(1 << ((sbi)->write_io_size_bits + 2)) /* KB */
@@ -114,6 +114,7 @@ struct f2fs_super_block {
 /*
  * For checkpoint
  */
+#define CP_TRIMMED_FLAG		0x00000100
 #define CP_NAT_BITS_FLAG	0x00000080
 #define CP_CRC_RECOVERY_FLAG	0x00000040
 #define CP_FASTBOOT_FLAG	0x00000020
@@ -161,7 +162,7 @@ struct f2fs_checkpoint {
  */
 #define F2FS_ORPHANS_PER_BLOCK	1020
 
-#define GET_ORPHAN_BLOCKS(n)	((n + F2FS_ORPHANS_PER_BLOCK - 1) / \
+#define GET_ORPHAN_BLOCKS(n)	(((n) + F2FS_ORPHANS_PER_BLOCK - 1) / \
 					F2FS_ORPHANS_PER_BLOCK)
 
 struct f2fs_orphan_block {
@@ -301,6 +302,12 @@ struct f2fs_nat_block {
 #define SIT_VBLOCK_MAP_SIZE 64
 #define SIT_ENTRY_PER_BLOCK (PAGE_SIZE / sizeof(struct f2fs_sit_entry))
 
+/*
+ * F2FS uses 4 bytes to represent block address. As a result, supported size of
+ * disk is 16 TB and it equals to 16 * 1024 * 1024 / 2 segments.
+ */
+#define F2FS_MAX_SEGMENT       ((16 * 1024 * 1024) / 2)
+
 /*
  * Note that f2fs_sit_entry->vblocks has the following bit-field information.
  * [15:10] : allocation type such as CURSEG_XXXX_TYPE
@@ -449,7 +456,7 @@ typedef __le32	f2fs_hash_t;
 #define F2FS_SLOT_LEN		8
 #define F2FS_SLOT_LEN_BITS	3
 
-#define GET_DENTRY_SLOTS(x)	((x + F2FS_SLOT_LEN - 1) >> F2FS_SLOT_LEN_BITS)
+#define GET_DENTRY_SLOTS(x) (((x) + F2FS_SLOT_LEN - 1) >> F2FS_SLOT_LEN_BITS)
 
 /* MAX level for dir lookup */
 #define MAX_DIR_HASH_DEPTH	63

include/trace/events/f2fs.h

@@ -15,6 +15,8 @@ TRACE_DEFINE_ENUM(META);
 TRACE_DEFINE_ENUM(META_FLUSH);
 TRACE_DEFINE_ENUM(INMEM);
 TRACE_DEFINE_ENUM(INMEM_DROP);
+TRACE_DEFINE_ENUM(INMEM_INVALIDATE);
+TRACE_DEFINE_ENUM(INMEM_REVOKE);
 TRACE_DEFINE_ENUM(IPU);
 TRACE_DEFINE_ENUM(OPU);
 TRACE_DEFINE_ENUM(CURSEG_HOT_DATA);
@@ -42,6 +44,7 @@ TRACE_DEFINE_ENUM(CP_FASTBOOT);
 TRACE_DEFINE_ENUM(CP_SYNC);
 TRACE_DEFINE_ENUM(CP_RECOVERY);
 TRACE_DEFINE_ENUM(CP_DISCARD);
+TRACE_DEFINE_ENUM(CP_TRIMMED);
 
 #define show_block_type(type)						\
 	__print_symbolic(type,						\
@@ -51,12 +54,13 @@ TRACE_DEFINE_ENUM(CP_DISCARD);
 		{ META_FLUSH,	"META_FLUSH" },				\
 		{ INMEM,	"INMEM" },				\
 		{ INMEM_DROP,	"INMEM_DROP" },				\
+		{ INMEM_INVALIDATE,	"INMEM_INVALIDATE" },		\
 		{ INMEM_REVOKE,	"INMEM_REVOKE" },			\
 		{ IPU,		"IN-PLACE" },				\
 		{ OPU,		"OUT-OF-PLACE" })
 
-#define F2FS_OP_FLAGS (REQ_RAHEAD | REQ_SYNC | REQ_PREFLUSH | REQ_META |\
-			REQ_PRIO)
+#define F2FS_OP_FLAGS (REQ_RAHEAD | REQ_SYNC | REQ_META | REQ_PRIO |	\
+			REQ_PREFLUSH | REQ_FUA)
 #define F2FS_BIO_FLAG_MASK(t)	(t & F2FS_OP_FLAGS)
 
 #define show_bio_type(op,op_flags)	show_bio_op(op),		\
@@ -75,16 +79,13 @@ TRACE_DEFINE_ENUM(CP_DISCARD);
 		{ REQ_OP_WRITE_ZEROES,		"WRITE_ZEROES" })
 
 #define show_bio_op_flags(flags)					\
-	__print_symbolic(F2FS_BIO_FLAG_MASK(flags),			\
-		{ REQ_RAHEAD, 		"(RA)" },			\
-		{ REQ_SYNC, 		"(S)" },			\
-		{ REQ_SYNC | REQ_PRIO,	"(SP)" },			\
-		{ REQ_META, 		"(M)" },			\
-		{ REQ_META | REQ_PRIO,	"(MP)" },			\
-		{ REQ_SYNC | REQ_PREFLUSH , "(SF)" },			\
-		{ REQ_SYNC | REQ_META | REQ_PRIO, "(SMP)" },		\
-		{ REQ_PREFLUSH | REQ_META | REQ_PRIO, "(FMP)" },	\
-		{ 0, " \b" })
+	__print_flags(F2FS_BIO_FLAG_MASK(flags), "|",			\
+		{ REQ_RAHEAD,		"R" },				\
+		{ REQ_SYNC,		"S" },				\
+		{ REQ_META,		"M" },				\
+		{ REQ_PRIO,		"P" },				\
+		{ REQ_PREFLUSH,		"PF" },				\
+		{ REQ_FUA,		"FUA" })
 
 #define show_data_type(type)						\
 	__print_symbolic(type,						\
@@ -117,12 +118,14 @@ TRACE_DEFINE_ENUM(CP_DISCARD);
 		{ GC_CB,	"Cost-Benefit" })
 
 #define show_cpreason(type)						\
-	__print_symbolic(type,						\
+	__print_flags(type, "|",					\
 		{ CP_UMOUNT,	"Umount" },				\
 		{ CP_FASTBOOT,	"Fastboot" },				\
 		{ CP_SYNC,	"Sync" },				\
 		{ CP_RECOVERY,	"Recovery" },				\
-		{ CP_DISCARD,	"Discard" })
+		{ CP_DISCARD,	"Discard" },				\
+		{ CP_UMOUNT,	"Umount" },				\
+		{ CP_TRIMMED,	"Trimmed" })
 
 struct victim_sel_policy;
 struct f2fs_map_blocks;
@@ -769,7 +772,7 @@ DECLARE_EVENT_CLASS(f2fs__submit_page_bio,
 	),
 
 	TP_printk("dev = (%d,%d), ino = %lu, page_index = 0x%lx, "
-		"oldaddr = 0x%llx, newaddr = 0x%llx, rw = %s%s, type = %s",
+		"oldaddr = 0x%llx, newaddr = 0x%llx, rw = %s(%s), type = %s",
 		show_dev_ino(__entry),
 		(unsigned long)__entry->index,
 		(unsigned long long)__entry->old_blkaddr,
@@ -822,7 +825,7 @@ DECLARE_EVENT_CLASS(f2fs__bio,
 		__entry->size		= bio->bi_iter.bi_size;
 	),
 
-	TP_printk("dev = (%d,%d)/(%d,%d), rw = %s%s, %s, sector = %lld, size = %u",
+	TP_printk("dev = (%d,%d)/(%d,%d), rw = %s(%s), %s, sector = %lld, size = %u",
 		show_dev(__entry->target),
 		show_dev(__entry->dev),
 		show_bio_type(__entry->op, __entry->op_flags),
@@ -1126,7 +1129,7 @@ TRACE_EVENT(f2fs_write_checkpoint,
 		__entry->msg)
 );
 
-TRACE_EVENT(f2fs_issue_discard,
+DECLARE_EVENT_CLASS(f2fs_discard,
 
 	TP_PROTO(struct block_device *dev, block_t blkstart, block_t blklen),
 
@@ -1150,6 +1153,20 @@ TRACE_EVENT(f2fs_issue_discard,
 		(unsigned long long)__entry->blklen)
 );
 
+DEFINE_EVENT(f2fs_discard, f2fs_queue_discard,
+
+	TP_PROTO(struct block_device *dev, block_t blkstart, block_t blklen),
+
+	TP_ARGS(dev, blkstart, blklen)
+);
+
+DEFINE_EVENT(f2fs_discard, f2fs_issue_discard,
+
+	TP_PROTO(struct block_device *dev, block_t blkstart, block_t blklen),
+
+	TP_ARGS(dev, blkstart, blklen)
+);
+
 TRACE_EVENT(f2fs_issue_reset_zone,
 
 	TP_PROTO(struct block_device *dev, block_t blkstart),
@@ -1174,26 +1191,29 @@ TRACE_EVENT(f2fs_issue_reset_zone,
 TRACE_EVENT(f2fs_issue_flush,
 
 	TP_PROTO(struct block_device *dev, unsigned int nobarrier,
-					unsigned int flush_merge),
+				unsigned int flush_merge, int ret),
 
-	TP_ARGS(dev, nobarrier, flush_merge),
+	TP_ARGS(dev, nobarrier, flush_merge, ret),
 
 	TP_STRUCT__entry(
 		__field(dev_t,	dev)
 		__field(unsigned int, nobarrier)
 		__field(unsigned int, flush_merge)
+		__field(int,  ret)
 	),
 
 	TP_fast_assign(
 		__entry->dev	= dev->bd_dev;
 		__entry->nobarrier = nobarrier;
 		__entry->flush_merge = flush_merge;
+		__entry->ret = ret;
 	),
 
-	TP_printk("dev = (%d,%d), %s %s",
+	TP_printk("dev = (%d,%d), %s %s, ret = %d",
 		show_dev(__entry->dev),
 		__entry->nobarrier ? "skip (nobarrier)" : "issue",
-		__entry->flush_merge ? " with flush_merge" : "")
+		__entry->flush_merge ? " with flush_merge" : "",
+		__entry->ret)
 );
 
 TRACE_EVENT(f2fs_lookup_extent_tree_start,