Merge tag 'for-f2fs-4.10' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "This patch series contains several performance tuning patches
  regarding to the IO submission flow, in addition to supporting new
  features such as a ZBC-base drive and multiple devices.

  It also includes some major bug fixes such as:
   - checkpoint version control
   - fdatasync-related roll-forward recovery routine
   - memory boundary or null-pointer access in corner cases
   - missing error cases

  It has various minor clean-up patches as well"

* tag 'for-f2fs-4.10' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (66 commits)
  f2fs: fix a missing size change in f2fs_setattr
  f2fs: fix to access nullified flush_cmd_control pointer
  f2fs: free meta pages if sanity check for ckpt is failed
  f2fs: detect wrong layout
  f2fs: call sync_fs when f2fs is idle
  Revert "f2fs: use percpu_counter for # of dirty pages in inode"
  f2fs: return AOP_WRITEPAGE_ACTIVATE for writepage
  f2fs: do not activate auto_recovery for fallocated i_size
  f2fs: fix to determine start_cp_addr by sbi->cur_cp_pack
  f2fs: fix 32-bit build
  f2fs: set ->owner for debugfs status file's file_operations
  f2fs: fix incorrect free inode count in ->statfs
  f2fs: drop duplicate header timer.h
  f2fs: fix wrong AUTO_RECOVER condition
  f2fs: do not recover i_size if it's valid
  f2fs: fix fdatasync
  f2fs: fix to account total free nid correctly
  f2fs: fix an infinite loop when flush nodes in cp
  f2fs: don't wait writeback for datas during checkpoint
  f2fs: fix wrong written_valid_blocks counting
  ...

1 files changed, 141 insertions, 85 deletions

diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index d1e29deb4598..b9078fdb3743 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -45,8 +45,8 @@ bool available_free_memory(struct f2fs_sb_info *sbi, int type)
 	 * give 25%, 25%, 50%, 50%, 50% memory for each components respectively
 	 */
 	if (type == FREE_NIDS) {
-		mem_size = (nm_i->fcnt * sizeof(struct free_nid)) >>
-							PAGE_SHIFT;
+		mem_size = (nm_i->nid_cnt[FREE_NID_LIST] *
+				sizeof(struct free_nid)) >> PAGE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 2);
 	} else if (type == NAT_ENTRIES) {
 		mem_size = (nm_i->nat_cnt * sizeof(struct nat_entry)) >>
@@ -270,8 +270,9 @@ static void cache_nat_entry(struct f2fs_sb_info *sbi, nid_t nid,
 		e = grab_nat_entry(nm_i, nid);
 		node_info_from_raw_nat(&e->ni, ne);
 	} else {
-		f2fs_bug_on(sbi, nat_get_ino(e) != ne->ino ||
-				nat_get_blkaddr(e) != ne->block_addr ||
+		f2fs_bug_on(sbi, nat_get_ino(e) != le32_to_cpu(ne->ino) ||
+				nat_get_blkaddr(e) !=
+					le32_to_cpu(ne->block_addr) ||
 				nat_get_version(e) != ne->version);
 	}
 }
@@ -1204,6 +1205,7 @@ static void flush_inline_data(struct f2fs_sb_info *sbi, nid_t ino)
 
 	ret = f2fs_write_inline_data(inode, page);
 	inode_dec_dirty_pages(inode);
+	remove_dirty_inode(inode);
 	if (ret)
 		set_page_dirty(page);
 page_out:
@@ -1338,7 +1340,8 @@ retry:
 			if (unlikely(f2fs_cp_error(sbi))) {
 				f2fs_put_page(last_page, 0);
 				pagevec_release(&pvec);
-				return -EIO;
+				ret = -EIO;
+				goto out;
 			}
 
 			if (!IS_DNODE(page) || !is_cold_node(page))
@@ -1407,11 +1410,12 @@ continue_unlock:
 			"Retry to write fsync mark: ino=%u, idx=%lx",
 					ino, last_page->index);
 		lock_page(last_page);
+		f2fs_wait_on_page_writeback(last_page, NODE, true);
 		set_page_dirty(last_page);
 		unlock_page(last_page);
 		goto retry;
 	}
-
+out:
 	if (nwritten)
 		f2fs_submit_merged_bio_cond(sbi, NULL, NULL, ino, NODE, WRITE);
 	return ret ? -EIO: 0;
@@ -1692,11 +1696,35 @@ static struct free_nid *__lookup_free_nid_list(struct f2fs_nm_info *nm_i,
 	return radix_tree_lookup(&nm_i->free_nid_root, n);
 }
 
-static void __del_from_free_nid_list(struct f2fs_nm_info *nm_i,
-						struct free_nid *i)
+static int __insert_nid_to_list(struct f2fs_sb_info *sbi,
+			struct free_nid *i, enum nid_list list, bool new)
 {
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+
+	if (new) {
+		int err = radix_tree_insert(&nm_i->free_nid_root, i->nid, i);
+		if (err)
+			return err;
+	}
+
+	f2fs_bug_on(sbi, list == FREE_NID_LIST ? i->state != NID_NEW :
+						i->state != NID_ALLOC);
+	nm_i->nid_cnt[list]++;
+	list_add_tail(&i->list, &nm_i->nid_list[list]);
+	return 0;
+}
+
+static void __remove_nid_from_list(struct f2fs_sb_info *sbi,
+			struct free_nid *i, enum nid_list list, bool reuse)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+
+	f2fs_bug_on(sbi, list == FREE_NID_LIST ? i->state != NID_NEW :
+						i->state != NID_ALLOC);
+	nm_i->nid_cnt[list]--;
 	list_del(&i->list);
-	radix_tree_delete(&nm_i->free_nid_root, i->nid);
+	if (!reuse)
+		radix_tree_delete(&nm_i->free_nid_root, i->nid);
 }
 
 static int add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
@@ -1704,9 +1732,7 @@ static int 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 nat_entry *ne;
-
-	if (!available_free_memory(sbi, FREE_NIDS))
-		return -1;
+	int err;
 
 	/* 0 nid should not be used */
 	if (unlikely(nid == 0))
@@ -1729,33 +1755,30 @@ static int add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
 		return 0;
 	}
 
-	spin_lock(&nm_i->free_nid_list_lock);
-	if (radix_tree_insert(&nm_i->free_nid_root, i->nid, i)) {
-		spin_unlock(&nm_i->free_nid_list_lock);
-		radix_tree_preload_end();
+	spin_lock(&nm_i->nid_list_lock);
+	err = __insert_nid_to_list(sbi, i, FREE_NID_LIST, true);
+	spin_unlock(&nm_i->nid_list_lock);
+	radix_tree_preload_end();
+	if (err) {
 		kmem_cache_free(free_nid_slab, i);
 		return 0;
 	}
-	list_add_tail(&i->list, &nm_i->free_nid_list);
-	nm_i->fcnt++;
-	spin_unlock(&nm_i->free_nid_list_lock);
-	radix_tree_preload_end();
 	return 1;
 }
 
-static void remove_free_nid(struct f2fs_nm_info *nm_i, nid_t nid)
+static void remove_free_nid(struct f2fs_sb_info *sbi, nid_t nid)
 {
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct free_nid *i;
 	bool need_free = false;
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
 	i = __lookup_free_nid_list(nm_i, nid);
 	if (i && i->state == NID_NEW) {
-		__del_from_free_nid_list(nm_i, i);
-		nm_i->fcnt--;
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, false);
 		need_free = true;
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+	spin_unlock(&nm_i->nid_list_lock);
 
 	if (need_free)
 		kmem_cache_free(free_nid_slab, i);
@@ -1778,14 +1801,12 @@ static void scan_nat_page(struct f2fs_sb_info *sbi,
 
 		blk_addr = le32_to_cpu(nat_blk->entries[i].block_addr);
 		f2fs_bug_on(sbi, blk_addr == NEW_ADDR);
-		if (blk_addr == NULL_ADDR) {
-			if (add_free_nid(sbi, start_nid, true) < 0)
-				break;
-		}
+		if (blk_addr == NULL_ADDR)
+			add_free_nid(sbi, start_nid, true);
 	}
 }
 
-void build_free_nids(struct f2fs_sb_info *sbi)
+static void __build_free_nids(struct f2fs_sb_info *sbi, bool sync)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
@@ -1794,7 +1815,10 @@ void build_free_nids(struct f2fs_sb_info *sbi)
 	nid_t nid = nm_i->next_scan_nid;
 
 	/* Enough entries */
-	if (nm_i->fcnt >= NAT_ENTRY_PER_BLOCK)
+	if (nm_i->nid_cnt[FREE_NID_LIST] >= NAT_ENTRY_PER_BLOCK)
+		return;
+
+	if (!sync && !available_free_memory(sbi, FREE_NIDS))
 		return;
 
 	/* readahead nat pages to be scanned */
@@ -1830,7 +1854,7 @@ void build_free_nids(struct f2fs_sb_info *sbi)
 		if (addr == NULL_ADDR)
 			add_free_nid(sbi, nid, true);
 		else
-			remove_free_nid(nm_i, nid);
+			remove_free_nid(sbi, nid);
 	}
 	up_read(&curseg->journal_rwsem);
 	up_read(&nm_i->nat_tree_lock);
@@ -1839,6 +1863,13 @@ void build_free_nids(struct f2fs_sb_info *sbi)
 					nm_i->ra_nid_pages, META_NAT, false);
 }
 
+void build_free_nids(struct f2fs_sb_info *sbi, bool sync)
+{
+	mutex_lock(&NM_I(sbi)->build_lock);
+	__build_free_nids(sbi, sync);
+	mutex_unlock(&NM_I(sbi)->build_lock);
+}
+
 /*
  * If this function returns success, caller can obtain a new nid
  * from second parameter of this function.
@@ -1853,31 +1884,31 @@ retry:
 	if (time_to_inject(sbi, FAULT_ALLOC_NID))
 		return false;
 #endif
-	if (unlikely(sbi->total_valid_node_count + 1 > nm_i->available_nids))
-		return false;
+	spin_lock(&nm_i->nid_list_lock);
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	if (unlikely(nm_i->available_nids == 0)) {
+		spin_unlock(&nm_i->nid_list_lock);
+		return false;
+	}
 
 	/* We should not use stale free nids created by build_free_nids */
-	if (nm_i->fcnt && !on_build_free_nids(nm_i)) {
-		f2fs_bug_on(sbi, list_empty(&nm_i->free_nid_list));
-		list_for_each_entry(i, &nm_i->free_nid_list, list)
-			if (i->state == NID_NEW)
-				break;
-
-		f2fs_bug_on(sbi, i->state != NID_NEW);
+	if (nm_i->nid_cnt[FREE_NID_LIST] && !on_build_free_nids(nm_i)) {
+		f2fs_bug_on(sbi, list_empty(&nm_i->nid_list[FREE_NID_LIST]));
+		i = list_first_entry(&nm_i->nid_list[FREE_NID_LIST],
+					struct free_nid, list);
 		*nid = i->nid;
+
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, true);
 		i->state = NID_ALLOC;
-		nm_i->fcnt--;
-		spin_unlock(&nm_i->free_nid_list_lock);
+		__insert_nid_to_list(sbi, i, ALLOC_NID_LIST, false);
+		nm_i->available_nids--;
+		spin_unlock(&nm_i->nid_list_lock);
 		return true;
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+	spin_unlock(&nm_i->nid_list_lock);
 
 	/* Let's scan nat pages and its caches to get free nids */
-	mutex_lock(&nm_i->build_lock);
-	build_free_nids(sbi);
-	mutex_unlock(&nm_i->build_lock);
+	build_free_nids(sbi, true);
 	goto retry;
 }
 
@@ -1889,11 +1920,11 @@ void alloc_nid_done(struct f2fs_sb_info *sbi, nid_t nid)
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct free_nid *i;
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
 	i = __lookup_free_nid_list(nm_i, nid);
-	f2fs_bug_on(sbi, !i || i->state != NID_ALLOC);
-	__del_from_free_nid_list(nm_i, i);
-	spin_unlock(&nm_i->free_nid_list_lock);
+	f2fs_bug_on(sbi, !i);
+	__remove_nid_from_list(sbi, i, ALLOC_NID_LIST, false);
+	spin_unlock(&nm_i->nid_list_lock);
 
 	kmem_cache_free(free_nid_slab, i);
 }
@@ -1910,17 +1941,22 @@ void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid)
 	if (!nid)
 		return;
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
 	i = __lookup_free_nid_list(nm_i, nid);
-	f2fs_bug_on(sbi, !i || i->state != NID_ALLOC);
+	f2fs_bug_on(sbi, !i);
+
 	if (!available_free_memory(sbi, FREE_NIDS)) {
-		__del_from_free_nid_list(nm_i, i);
+		__remove_nid_from_list(sbi, i, ALLOC_NID_LIST, false);
 		need_free = true;
 	} else {
+		__remove_nid_from_list(sbi, i, ALLOC_NID_LIST, true);
 		i->state = NID_NEW;
-		nm_i->fcnt++;
+		__insert_nid_to_list(sbi, i, FREE_NID_LIST, false);
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+
+	nm_i->available_nids++;
+
+	spin_unlock(&nm_i->nid_list_lock);
 
 	if (need_free)
 		kmem_cache_free(free_nid_slab, i);
@@ -1932,24 +1968,24 @@ int try_to_free_nids(struct f2fs_sb_info *sbi, int nr_shrink)
 	struct free_nid *i, *next;
 	int nr = nr_shrink;
 
-	if (nm_i->fcnt <= MAX_FREE_NIDS)
+	if (nm_i->nid_cnt[FREE_NID_LIST] <= MAX_FREE_NIDS)
 		return 0;
 
 	if (!mutex_trylock(&nm_i->build_lock))
 		return 0;
 
-	spin_lock(&nm_i->free_nid_list_lock);
-	list_for_each_entry_safe(i, next, &nm_i->free_nid_list, list) {
-		if (nr_shrink <= 0 || nm_i->fcnt <= MAX_FREE_NIDS)
+	spin_lock(&nm_i->nid_list_lock);
+	list_for_each_entry_safe(i, next, &nm_i->nid_list[FREE_NID_LIST],
+									list) {
+		if (nr_shrink <= 0 ||
+				nm_i->nid_cnt[FREE_NID_LIST] <= MAX_FREE_NIDS)
 			break;
-		if (i->state == NID_ALLOC)
-			continue;
-		__del_from_free_nid_list(nm_i, i);
+
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, false);
 		kmem_cache_free(free_nid_slab, i);
-		nm_i->fcnt--;
 		nr_shrink--;
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+	spin_unlock(&nm_i->nid_list_lock);
 	mutex_unlock(&nm_i->build_lock);
 
 	return nr - nr_shrink;
@@ -2005,7 +2041,7 @@ recover_xnid:
 	if (unlikely(!inc_valid_node_count(sbi, inode)))
 		f2fs_bug_on(sbi, 1);
 
-	remove_free_nid(NM_I(sbi), new_xnid);
+	remove_free_nid(sbi, new_xnid);
 	get_node_info(sbi, new_xnid, &ni);
 	ni.ino = inode->i_ino;
 	set_node_addr(sbi, &ni, NEW_ADDR, false);
@@ -2035,7 +2071,7 @@ retry:
 	}
 
 	/* Should not use this inode from free nid list */
-	remove_free_nid(NM_I(sbi), ino);
+	remove_free_nid(sbi, ino);
 
 	if (!PageUptodate(ipage))
 		SetPageUptodate(ipage);
@@ -2069,7 +2105,6 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
 	struct f2fs_node *rn;
 	struct f2fs_summary *sum_entry;
 	block_t addr;
-	int bio_blocks = MAX_BIO_BLOCKS(sbi);
 	int i, idx, last_offset, nrpages;
 
 	/* scan the node segment */
@@ -2078,7 +2113,7 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
 	sum_entry = &sum->entries[0];
 
 	for (i = 0; i < last_offset; i += nrpages, addr += nrpages) {
-		nrpages = min(last_offset - i, bio_blocks);
+		nrpages = min(last_offset - i, BIO_MAX_PAGES);
 
 		/* readahead node pages */
 		ra_meta_pages(sbi, addr, nrpages, META_POR, true);
@@ -2120,6 +2155,19 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
 			ne = grab_nat_entry(nm_i, nid);
 			node_info_from_raw_nat(&ne->ni, &raw_ne);
 		}
+
+		/*
+		 * if a free nat in journal has not been used after last
+		 * checkpoint, we should remove it from available nids,
+		 * since later we will add it again.
+		 */
+		if (!get_nat_flag(ne, IS_DIRTY) &&
+				le32_to_cpu(raw_ne.block_addr) == NULL_ADDR) {
+			spin_lock(&nm_i->nid_list_lock);
+			nm_i->available_nids--;
+			spin_unlock(&nm_i->nid_list_lock);
+		}
+
 		__set_nat_cache_dirty(nm_i, ne);
 	}
 	update_nats_in_cursum(journal, -i);
@@ -2192,8 +2240,12 @@ 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);
-		if (nat_get_blkaddr(ne) == NULL_ADDR)
+		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++;
+			spin_unlock(&NM_I(sbi)->nid_list_lock);
+		}
 	}
 
 	if (to_journal)
@@ -2268,21 +2320,24 @@ 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 - F2FS_RESERVED_NODE_NUM;
-	nm_i->fcnt = 0;
+	nm_i->available_nids = nm_i->max_nid - sbi->total_valid_node_count -
+							F2FS_RESERVED_NODE_NUM;
+	nm_i->nid_cnt[FREE_NID_LIST] = 0;
+	nm_i->nid_cnt[ALLOC_NID_LIST] = 0;
 	nm_i->nat_cnt = 0;
 	nm_i->ram_thresh = DEF_RAM_THRESHOLD;
 	nm_i->ra_nid_pages = DEF_RA_NID_PAGES;
 	nm_i->dirty_nats_ratio = DEF_DIRTY_NAT_RATIO_THRESHOLD;
 
 	INIT_RADIX_TREE(&nm_i->free_nid_root, GFP_ATOMIC);
-	INIT_LIST_HEAD(&nm_i->free_nid_list);
+	INIT_LIST_HEAD(&nm_i->nid_list[FREE_NID_LIST]);
+	INIT_LIST_HEAD(&nm_i->nid_list[ALLOC_NID_LIST]);
 	INIT_RADIX_TREE(&nm_i->nat_root, GFP_NOIO);
 	INIT_RADIX_TREE(&nm_i->nat_set_root, GFP_NOIO);
 	INIT_LIST_HEAD(&nm_i->nat_entries);
 
 	mutex_init(&nm_i->build_lock);
-	spin_lock_init(&nm_i->free_nid_list_lock);
+	spin_lock_init(&nm_i->nid_list_lock);
 	init_rwsem(&nm_i->nat_tree_lock);
 
 	nm_i->next_scan_nid = le32_to_cpu(sbi->ckpt->next_free_nid);
@@ -2310,7 +2365,7 @@ int build_node_manager(struct f2fs_sb_info *sbi)
 	if (err)
 		return err;
 
-	build_free_nids(sbi);
+	build_free_nids(sbi, true);
 	return 0;
 }
 
@@ -2327,17 +2382,18 @@ void destroy_node_manager(struct f2fs_sb_info *sbi)
 		return;
 
 	/* destroy free nid list */
-	spin_lock(&nm_i->free_nid_list_lock);
-	list_for_each_entry_safe(i, next_i, &nm_i->free_nid_list, list) {
-		f2fs_bug_on(sbi, i->state == NID_ALLOC);
-		__del_from_free_nid_list(nm_i, i);
-		nm_i->fcnt--;
-		spin_unlock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
+	list_for_each_entry_safe(i, next_i, &nm_i->nid_list[FREE_NID_LIST],
+									list) {
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, false);
+		spin_unlock(&nm_i->nid_list_lock);
 		kmem_cache_free(free_nid_slab, i);
-		spin_lock(&nm_i->free_nid_list_lock);
+		spin_lock(&nm_i->nid_list_lock);
 	}
-	f2fs_bug_on(sbi, nm_i->fcnt);
-	spin_unlock(&nm_i->free_nid_list_lock);
+	f2fs_bug_on(sbi, nm_i->nid_cnt[FREE_NID_LIST]);
+	f2fs_bug_on(sbi, nm_i->nid_cnt[ALLOC_NID_LIST]);
+	f2fs_bug_on(sbi, !list_empty(&nm_i->nid_list[ALLOC_NID_LIST]));
+	spin_unlock(&nm_i->nid_list_lock);
 
 	/* destroy nat cache */
 	down_write(&nm_i->nat_tree_lock);