Merge tag 'f2fs-for-4.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "In this round, we've followed up to support some generic features such
  as cgroup, block reservation, linking fscrypt_ops, delivering
  write_hints, and some ioctls. And, we could fix some corner cases in
  terms of power-cut recovery and subtle deadlocks.

  Enhancements:
   - bitmap operations to handle NAT blocks
   - readahead to improve readdir speed
   - switch to use fscrypt_*
   - apply write hints for direct IO
   - add reserve_root=%u,resuid=%u,resgid=%u to reserve blocks for root/uid/gid
   - modify b_avail and b_free to consider root reserved blocks
   - support cgroup writeback
   - support FIEMAP_FLAG_XATTR for fibmap
   - add F2FS_IOC_PRECACHE_EXTENTS to pre-cache extents
   - add F2FS_IOC_{GET/SET}_PIN_FILE to pin LBAs for data blocks
   - support inode creation time

  Bug fixs:
   - sysfile-based quota operations
   - memory footprint accounting
   - allow to write data on partial preallocation case
   - fix deadlock case on fallocate
   - fix to handle fill_super errors
   - fix missing inode updates of fsync'ed file
   - recover renamed file which was fsycn'ed before
   - drop inmemory pages in corner error case
   - keep last_disk_size correctly
   - recover missing i_inline flags during roll-forward

  Various clean-up patches were added as well"

* tag 'f2fs-for-4.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (72 commits)
  f2fs: support inode creation time
  f2fs: rebuild sit page from sit info in mem
  f2fs: stop issuing discard if fs is readonly
  f2fs: clean up duplicated assignment in init_discard_policy
  f2fs: use GFP_F2FS_ZERO for cleanup
  f2fs: allow to recover node blocks given updated checkpoint
  f2fs: recover some i_inline flags
  f2fs: correct removexattr behavior for null valued extended attribute
  f2fs: drop page cache after fs shutdown
  f2fs: stop gc/discard thread after fs shutdown
  f2fs: hanlde error case in f2fs_ioc_shutdown
  f2fs: split need_inplace_update
  f2fs: fix to update last_disk_size correctly
  f2fs: kill F2FS_INLINE_XATTR_ADDRS for cleanup
  f2fs: clean up error path of fill_super
  f2fs: avoid hungtask when GC encrypted block if io_bits is set
  f2fs: allow quota to use reserved blocks
  f2fs: fix to drop all inmem pages correctly
  f2fs: speed up defragment on sparse file
  f2fs: support F2FS_IOC_PRECACHE_EXTENTS
  ...

Documentation/ABI/testing/sysfs-fs-f2fs

@@ -186,3 +186,9 @@ Date:		August 2017
 Contact:	"Jaegeuk Kim" <jaegeuk@kernel.org>
 Description:
 		 Controls sleep time of GC urgent mode
+
+What:		/sys/fs/f2fs/<disk>/readdir_ra
+Date:		November 2017
+Contact:	"Sheng Yong" <shengyong1@huawei.com>
+Description:
+		 Controls readahead inode block in readdir.

fs/f2fs/acl.c

@@ -270,7 +270,7 @@ static struct posix_acl *f2fs_acl_clone(const struct posix_acl *acl,
 				sizeof(struct posix_acl_entry);
 		clone = kmemdup(acl, size, flags);
 		if (clone)
-			atomic_set(&clone->a_refcount, 1);
+			refcount_set(&clone->a_refcount, 1);
 	}
 	return clone;
 }

fs/f2fs/checkpoint.c

@@ -237,12 +237,15 @@ static int __f2fs_write_meta_page(struct page *page,
 
 	trace_f2fs_writepage(page, META);
 
+	if (unlikely(f2fs_cp_error(sbi))) {
+		dec_page_count(sbi, F2FS_DIRTY_META);
+		unlock_page(page);
+		return 0;
+	}
 	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
 		goto redirty_out;
 	if (wbc->for_reclaim && page->index < GET_SUM_BLOCK(sbi, 0))
 		goto redirty_out;
-	if (unlikely(f2fs_cp_error(sbi)))
-		goto redirty_out;
 
 	write_meta_page(sbi, page, io_type);
 	dec_page_count(sbi, F2FS_DIRTY_META);
@@ -793,7 +796,7 @@ int get_valid_checkpoint(struct f2fs_sb_info *sbi)
 	block_t cp_blk_no;
 	int i;
 
-	sbi->ckpt = kzalloc(cp_blks * blk_size, GFP_KERNEL);
+	sbi->ckpt = f2fs_kzalloc(sbi, cp_blks * blk_size, GFP_KERNEL);
 	if (!sbi->ckpt)
 		return -ENOMEM;
 	/*
@@ -1154,6 +1157,7 @@ static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	/* set this flag to activate crc|cp_ver for recovery */
 	__set_ckpt_flags(ckpt, CP_CRC_RECOVERY_FLAG);
+	__clear_ckpt_flags(ckpt, CP_NOCRC_RECOVERY_FLAG);
 
 	spin_unlock_irqrestore(&sbi->cp_lock, flags);
 }

fs/f2fs/data.c

@@ -111,8 +111,13 @@ static void f2fs_write_end_io(struct bio *bio)
 
 		if (unlikely(bio->bi_status)) {
 			mapping_set_error(page->mapping, -EIO);
-			f2fs_stop_checkpoint(sbi, true);
+			if (type == F2FS_WB_CP_DATA)
+				f2fs_stop_checkpoint(sbi, true);
 		}
+
+		f2fs_bug_on(sbi, page->mapping == NODE_MAPPING(sbi) &&
+					page->index != nid_of_node(page));
+
 		dec_page_count(sbi, type);
 		clear_cold_data(page);
 		end_page_writeback(page);
@@ -169,6 +174,7 @@ static bool __same_bdev(struct f2fs_sb_info *sbi,
  * Low-level block read/write IO operations.
  */
 static struct bio *__bio_alloc(struct f2fs_sb_info *sbi, block_t blk_addr,
+				struct writeback_control *wbc,
 				int npages, bool is_read)
 {
 	struct bio *bio;
@@ -178,6 +184,8 @@ static struct bio *__bio_alloc(struct f2fs_sb_info *sbi, block_t blk_addr,
 	f2fs_target_device(sbi, blk_addr, bio);
 	bio->bi_end_io = is_read ? f2fs_read_end_io : f2fs_write_end_io;
 	bio->bi_private = is_read ? NULL : sbi;
+	if (wbc)
+		wbc_init_bio(wbc, bio);
 
 	return bio;
 }
@@ -373,7 +381,8 @@ int f2fs_submit_page_bio(struct f2fs_io_info *fio)
 	f2fs_trace_ios(fio, 0);
 
 	/* Allocate a new bio */
-	bio = __bio_alloc(fio->sbi, fio->new_blkaddr, 1, is_read_io(fio->op));
+	bio = __bio_alloc(fio->sbi, fio->new_blkaddr, fio->io_wbc,
+				1, is_read_io(fio->op));
 
 	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
 		bio_put(bio);
@@ -435,7 +444,7 @@ alloc_new:
 			dec_page_count(sbi, WB_DATA_TYPE(bio_page));
 			goto out_fail;
 		}
-		io->bio = __bio_alloc(sbi, fio->new_blkaddr,
+		io->bio = __bio_alloc(sbi, fio->new_blkaddr, fio->io_wbc,
 						BIO_MAX_PAGES, false);
 		io->fio = *fio;
 	}
@@ -445,6 +454,9 @@ alloc_new:
 		goto alloc_new;
 	}
 
+	if (fio->io_wbc)
+		wbc_account_io(fio->io_wbc, bio_page, PAGE_SIZE);
+
 	io->last_block_in_bio = fio->new_blkaddr;
 	f2fs_trace_ios(fio, 0);
 
@@ -783,7 +795,7 @@ got_it:
 	return page;
 }
 
-static int __allocate_data_block(struct dnode_of_data *dn)
+static int __allocate_data_block(struct dnode_of_data *dn, int seg_type)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
 	struct f2fs_summary sum;
@@ -808,7 +820,7 @@ alloc:
 	set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
 
 	allocate_data_block(sbi, NULL, dn->data_blkaddr, &dn->data_blkaddr,
-					&sum, CURSEG_WARM_DATA, NULL, false);
+					&sum, seg_type, NULL, false);
 	set_data_blkaddr(dn);
 
 	/* update i_size */
@@ -831,10 +843,12 @@ int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from)
 {
 	struct inode *inode = file_inode(iocb->ki_filp);
 	struct f2fs_map_blocks map;
+	int flag;
 	int err = 0;
+	bool direct_io = iocb->ki_flags & IOCB_DIRECT;
 
 	/* convert inline data for Direct I/O*/
-	if (iocb->ki_flags & IOCB_DIRECT) {
+	if (direct_io) {
 		err = f2fs_convert_inline_inode(inode);
 		if (err)
 			return err;
@@ -851,19 +865,33 @@ int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from)
 		map.m_len = 0;
 
 	map.m_next_pgofs = NULL;
+	map.m_next_extent = NULL;
+	map.m_seg_type = NO_CHECK_TYPE;
 
-	if (iocb->ki_flags & IOCB_DIRECT)
-		return f2fs_map_blocks(inode, &map, 1,
-			__force_buffered_io(inode, WRITE) ?
-				F2FS_GET_BLOCK_PRE_AIO :
-				F2FS_GET_BLOCK_PRE_DIO);
+	if (direct_io) {
+		map.m_seg_type = rw_hint_to_seg_type(iocb->ki_hint);
+		flag = __force_buffered_io(inode, WRITE) ?
+					F2FS_GET_BLOCK_PRE_AIO :
+					F2FS_GET_BLOCK_PRE_DIO;
+		goto map_blocks;
+	}
 	if (iocb->ki_pos + iov_iter_count(from) > MAX_INLINE_DATA(inode)) {
 		err = f2fs_convert_inline_inode(inode);
 		if (err)
 			return err;
 	}
-	if (!f2fs_has_inline_data(inode))
-		return f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_AIO);
+	if (f2fs_has_inline_data(inode))
+		return err;
+
+	flag = F2FS_GET_BLOCK_PRE_AIO;
+
+map_blocks:
+	err = f2fs_map_blocks(inode, &map, 1, flag);
+	if (map.m_len > 0 && err == -ENOSPC) {
+		if (!direct_io)
+			set_inode_flag(inode, FI_NO_PREALLOC);
+		err = 0;
+	}
 	return err;
 }
 
@@ -904,6 +932,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
 	blkcnt_t prealloc;
 	struct extent_info ei = {0,0,0};
 	block_t blkaddr;
+	unsigned int start_pgofs;
 
 	if (!maxblocks)
 		return 0;
@@ -919,6 +948,8 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
 		map->m_pblk = ei.blk + pgofs - ei.fofs;
 		map->m_len = min((pgoff_t)maxblocks, ei.fofs + ei.len - pgofs);
 		map->m_flags = F2FS_MAP_MAPPED;
+		if (map->m_next_extent)
+			*map->m_next_extent = pgofs + map->m_len;
 		goto out;
 	}
 
@@ -937,10 +968,14 @@ next_dnode:
 			if (map->m_next_pgofs)
 				*map->m_next_pgofs =
 					get_next_page_offset(&dn, pgofs);
+			if (map->m_next_extent)
+				*map->m_next_extent =
+					get_next_page_offset(&dn, pgofs);
 		}
 		goto unlock_out;
 	}
 
+	start_pgofs = pgofs;
 	prealloc = 0;
 	last_ofs_in_node = ofs_in_node = dn.ofs_in_node;
 	end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
@@ -960,7 +995,8 @@ next_block:
 					last_ofs_in_node = dn.ofs_in_node;
 				}
 			} else {
-				err = __allocate_data_block(&dn);
+				err = __allocate_data_block(&dn,
+							map->m_seg_type);
 				if (!err)
 					set_inode_flag(inode, FI_APPEND_WRITE);
 			}
@@ -973,14 +1009,20 @@ next_block:
 				map->m_pblk = 0;
 				goto sync_out;
 			}
+			if (flag == F2FS_GET_BLOCK_PRECACHE)
+				goto sync_out;
 			if (flag == F2FS_GET_BLOCK_FIEMAP &&
 						blkaddr == NULL_ADDR) {
 				if (map->m_next_pgofs)
 					*map->m_next_pgofs = pgofs + 1;
+				goto sync_out;
 			}
-			if (flag != F2FS_GET_BLOCK_FIEMAP ||
-						blkaddr != NEW_ADDR)
+			if (flag != F2FS_GET_BLOCK_FIEMAP) {
+				/* for defragment case */
+				if (map->m_next_pgofs)
+					*map->m_next_pgofs = pgofs + 1;
 				goto sync_out;
+			}
 		}
 	}
 
@@ -1031,6 +1073,16 @@ skip:
 	else if (dn.ofs_in_node < end_offset)
 		goto next_block;
 
+	if (flag == F2FS_GET_BLOCK_PRECACHE) {
+		if (map->m_flags & F2FS_MAP_MAPPED) {
+			unsigned int ofs = start_pgofs - map->m_lblk;
+
+			f2fs_update_extent_cache_range(&dn,
+				start_pgofs, map->m_pblk + ofs,
+				map->m_len - ofs);
+		}
+	}
+
 	f2fs_put_dnode(&dn);
 
 	if (create) {
@@ -1040,6 +1092,17 @@ skip:
 	goto next_dnode;
 
 sync_out:
+	if (flag == F2FS_GET_BLOCK_PRECACHE) {
+		if (map->m_flags & F2FS_MAP_MAPPED) {
+			unsigned int ofs = start_pgofs - map->m_lblk;
+
+			f2fs_update_extent_cache_range(&dn,
+				start_pgofs, map->m_pblk + ofs,
+				map->m_len - ofs);
+		}
+		if (map->m_next_extent)
+			*map->m_next_extent = pgofs + 1;
+	}
 	f2fs_put_dnode(&dn);
 unlock_out:
 	if (create) {
@@ -1053,7 +1116,7 @@ out:
 
 static int __get_data_block(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh, int create, int flag,
-			pgoff_t *next_pgofs)
+			pgoff_t *next_pgofs, int seg_type)
 {
 	struct f2fs_map_blocks map;
 	int err;
@@ -1061,6 +1124,8 @@ static int __get_data_block(struct inode *inode, sector_t iblock,
 	map.m_lblk = iblock;
 	map.m_len = bh->b_size >> inode->i_blkbits;
 	map.m_next_pgofs = next_pgofs;
+	map.m_next_extent = NULL;
+	map.m_seg_type = seg_type;
 
 	err = f2fs_map_blocks(inode, &map, create, flag);
 	if (!err) {
@@ -1076,14 +1141,17 @@ static int get_data_block(struct inode *inode, sector_t iblock,
 			pgoff_t *next_pgofs)
 {
 	return __get_data_block(inode, iblock, bh_result, create,
-							flag, next_pgofs);
+							flag, next_pgofs,
+							NO_CHECK_TYPE);
 }
 
 static int get_data_block_dio(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh_result, int create)
 {
 	return __get_data_block(inode, iblock, bh_result, create,
-						F2FS_GET_BLOCK_DEFAULT, NULL);
+						F2FS_GET_BLOCK_DEFAULT, NULL,
+						rw_hint_to_seg_type(
+							inode->i_write_hint));
 }
 
 static int get_data_block_bmap(struct inode *inode, sector_t iblock,
@@ -1094,7 +1162,8 @@ static int get_data_block_bmap(struct inode *inode, sector_t iblock,
 		return -EFBIG;
 
 	return __get_data_block(inode, iblock, bh_result, create,
-						F2FS_GET_BLOCK_BMAP, NULL);
+						F2FS_GET_BLOCK_BMAP, NULL,
+						NO_CHECK_TYPE);
 }
 
 static inline sector_t logical_to_blk(struct inode *inode, loff_t offset)
@@ -1107,6 +1176,68 @@ static inline loff_t blk_to_logical(struct inode *inode, sector_t blk)
 	return (blk << inode->i_blkbits);
 }
 
+static int f2fs_xattr_fiemap(struct inode *inode,
+				struct fiemap_extent_info *fieinfo)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct page *page;
+	struct node_info ni;
+	__u64 phys = 0, len;
+	__u32 flags;
+	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
+	int err = 0;
+
+	if (f2fs_has_inline_xattr(inode)) {
+		int offset;
+
+		page = f2fs_grab_cache_page(NODE_MAPPING(sbi),
+						inode->i_ino, false);
+		if (!page)
+			return -ENOMEM;
+
+		get_node_info(sbi, inode->i_ino, &ni);
+
+		phys = (__u64)blk_to_logical(inode, ni.blk_addr);
+		offset = offsetof(struct f2fs_inode, i_addr) +
+					sizeof(__le32) * (DEF_ADDRS_PER_INODE -
+					get_inline_xattr_addrs(inode));
+
+		phys += offset;
+		len = inline_xattr_size(inode);
+
+		f2fs_put_page(page, 1);
+
+		flags = FIEMAP_EXTENT_DATA_INLINE | FIEMAP_EXTENT_NOT_ALIGNED;
+
+		if (!xnid)
+			flags |= FIEMAP_EXTENT_LAST;
+
+		err = fiemap_fill_next_extent(fieinfo, 0, phys, len, flags);
+		if (err || err == 1)
+			return err;
+	}
+
+	if (xnid) {
+		page = f2fs_grab_cache_page(NODE_MAPPING(sbi), xnid, false);
+		if (!page)
+			return -ENOMEM;
+
+		get_node_info(sbi, xnid, &ni);
+
+		phys = (__u64)blk_to_logical(inode, ni.blk_addr);
+		len = inode->i_sb->s_blocksize;
+
+		f2fs_put_page(page, 1);
+
+		flags = FIEMAP_EXTENT_LAST;
+	}
+
+	if (phys)
+		err = fiemap_fill_next_extent(fieinfo, 0, phys, len, flags);
+
+	return (err < 0 ? err : 0);
+}
+
 int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		u64 start, u64 len)
 {
@@ -1117,18 +1248,29 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	u32 flags = 0;
 	int ret = 0;
 
-	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC);
+	if (fieinfo->fi_flags & FIEMAP_FLAG_CACHE) {
+		ret = f2fs_precache_extents(inode);
+		if (ret)
+			return ret;
+	}
+
+	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC | FIEMAP_FLAG_XATTR);
 	if (ret)
 		return ret;
 
+	inode_lock(inode);
+
+	if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) {
+		ret = f2fs_xattr_fiemap(inode, fieinfo);
+		goto out;
+	}
+
 	if (f2fs_has_inline_data(inode)) {
 		ret = f2fs_inline_data_fiemap(inode, fieinfo, start, len);
 		if (ret != -EAGAIN)
-			return ret;
+			goto out;
 	}
 
-	inode_lock(inode);
-
 	if (logical_to_blk(inode, len) == 0)
 		len = blk_to_logical(inode, 1);
 
@@ -1198,7 +1340,6 @@ static int f2fs_mpage_readpages(struct address_space *mapping,
 			unsigned nr_pages)
 {
 	struct bio *bio = NULL;
-	unsigned page_idx;
 	sector_t last_block_in_bio = 0;
 	struct inode *inode = mapping->host;
 	const unsigned blkbits = inode->i_blkbits;
@@ -1214,9 +1355,10 @@ static int f2fs_mpage_readpages(struct address_space *mapping,
 	map.m_len = 0;
 	map.m_flags = 0;
 	map.m_next_pgofs = NULL;
+	map.m_next_extent = NULL;
+	map.m_seg_type = NO_CHECK_TYPE;
 
-	for (page_idx = 0; nr_pages; page_idx++, nr_pages--) {
-
+	for (; nr_pages; nr_pages--) {
 		if (pages) {
 			page = list_last_entry(pages, struct page, lru);
 
@@ -1376,18 +1518,79 @@ retry_encrypt:
 	return PTR_ERR(fio->encrypted_page);
 }
 
+static inline bool check_inplace_update_policy(struct inode *inode,
+				struct f2fs_io_info *fio)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	unsigned int policy = SM_I(sbi)->ipu_policy;
+
+	if (policy & (0x1 << F2FS_IPU_FORCE))
+		return true;
+	if (policy & (0x1 << F2FS_IPU_SSR) && need_SSR(sbi))
+		return true;
+	if (policy & (0x1 << F2FS_IPU_UTIL) &&
+			utilization(sbi) > SM_I(sbi)->min_ipu_util)
+		return true;
+	if (policy & (0x1 << F2FS_IPU_SSR_UTIL) && need_SSR(sbi) &&
+			utilization(sbi) > SM_I(sbi)->min_ipu_util)
+		return true;
+
+	/*
+	 * IPU for rewrite async pages
+	 */
+	if (policy & (0x1 << F2FS_IPU_ASYNC) &&
+			fio && fio->op == REQ_OP_WRITE &&
+			!(fio->op_flags & REQ_SYNC) &&
+			!f2fs_encrypted_inode(inode))
+		return true;
+
+	/* this is only set during fdatasync */
+	if (policy & (0x1 << F2FS_IPU_FSYNC) &&
+			is_inode_flag_set(inode, FI_NEED_IPU))
+		return true;
+
+	return false;
+}
+
+bool should_update_inplace(struct inode *inode, struct f2fs_io_info *fio)
+{
+	if (f2fs_is_pinned_file(inode))
+		return true;
+
+	/* if this is cold file, we should overwrite to avoid fragmentation */
+	if (file_is_cold(inode))
+		return true;
+
+	return check_inplace_update_policy(inode, fio);
+}
+
+bool should_update_outplace(struct inode *inode, struct f2fs_io_info *fio)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+
+	if (test_opt(sbi, LFS))
+		return true;
+	if (S_ISDIR(inode->i_mode))
+		return true;
+	if (f2fs_is_atomic_file(inode))
+		return true;
+	if (fio) {
+		if (is_cold_data(fio->page))
+			return true;
+		if (IS_ATOMIC_WRITTEN_PAGE(fio->page))
+			return true;
+	}
+	return false;
+}
+
 static inline bool need_inplace_update(struct f2fs_io_info *fio)
 {
 	struct inode *inode = fio->page->mapping->host;
 
-	if (S_ISDIR(inode->i_mode) || f2fs_is_atomic_file(inode))
-		return false;
-	if (is_cold_data(fio->page))
-		return false;
-	if (IS_ATOMIC_WRITTEN_PAGE(fio->page))
+	if (should_update_outplace(inode, fio))
 		return false;
 
-	return need_inplace_update_policy(inode, fio);
+	return should_update_inplace(inode, fio);
 }
 
 static inline bool valid_ipu_blkaddr(struct f2fs_io_info *fio)
@@ -1508,10 +1711,17 @@ static int __write_data_page(struct page *page, bool *submitted,
 		.submitted = false,
 		.need_lock = LOCK_RETRY,
 		.io_type = io_type,
+		.io_wbc = wbc,
 	};
 
 	trace_f2fs_writepage(page, DATA);
 
+	/* we should bypass data pages to proceed the kworkder jobs */
+	if (unlikely(f2fs_cp_error(sbi))) {
+		mapping_set_error(page->mapping, -EIO);
+		goto out;
+	}
+
 	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
 		goto redirty_out;
 
@@ -1536,12 +1746,6 @@ write:
 			available_free_memory(sbi, BASE_CHECK))))
 		goto redirty_out;
 
-	/* we should bypass data pages to proceed the kworkder jobs */
-	if (unlikely(f2fs_cp_error(sbi))) {
-		mapping_set_error(page->mapping, -EIO);
-		goto out;
-	}
-
 	/* Dentry blocks are controlled by checkpoint */
 	if (S_ISDIR(inode->i_mode)) {
 		fio.need_lock = LOCK_DONE;
@@ -1571,10 +1775,14 @@ write:
 		}
 	}
 
-	down_write(&F2FS_I(inode)->i_sem);
-	if (F2FS_I(inode)->last_disk_size < psize)
-		F2FS_I(inode)->last_disk_size = psize;
-	up_write(&F2FS_I(inode)->i_sem);
+	if (err) {
+		file_set_keep_isize(inode);
+	} else {
+		down_write(&F2FS_I(inode)->i_sem);
+		if (F2FS_I(inode)->last_disk_size < psize)
+			F2FS_I(inode)->last_disk_size = psize;
+		up_write(&F2FS_I(inode)->i_sem);
+	}
 
 done:
 	if (err && err != -ENOENT)
@@ -1933,7 +2141,7 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct page *page = NULL;
 	pgoff_t index = ((unsigned long long) pos) >> PAGE_SHIFT;
-	bool need_balance = false;
+	bool need_balance = false, drop_atomic = false;
 	block_t blkaddr = NULL_ADDR;
 	int err = 0;
 
@@ -1942,6 +2150,7 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 	if (f2fs_is_atomic_file(inode) &&
 			!available_free_memory(sbi, INMEM_PAGES)) {
 		err = -ENOMEM;
+		drop_atomic = true;
 		goto fail;
 	}
 
@@ -2022,7 +2231,7 @@ repeat:
 fail:
 	f2fs_put_page(page, 1);
 	f2fs_write_failed(mapping, pos + len);
-	if (f2fs_is_atomic_file(inode))
+	if (drop_atomic)
 		drop_inmem_pages_all(sbi);
 	return err;
 }

fs/f2fs/debug.c

@@ -49,14 +49,7 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->ndirty_imeta = get_pages(sbi, F2FS_DIRTY_IMETA);
 	si->ndirty_dirs = sbi->ndirty_inode[DIR_INODE];
 	si->ndirty_files = sbi->ndirty_inode[FILE_INODE];
-
-	si->nquota_files = 0;
-	if (f2fs_sb_has_quota_ino(sbi->sb)) {
-		for (i = 0; i < MAXQUOTAS; i++) {
-			if (f2fs_qf_ino(sbi->sb, i))
-				si->nquota_files++;
-		}
-	}
+	si->nquota_files = sbi->nquota_files;
 	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);
@@ -186,7 +179,6 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	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;
 
 	/* build sm */
 	si->base_mem += sizeof(struct f2fs_sm_info);
@@ -447,7 +439,7 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi)
 	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
 	struct f2fs_stat_info *si;
 
-	si = kzalloc(sizeof(struct f2fs_stat_info), GFP_KERNEL);
+	si = f2fs_kzalloc(sbi, sizeof(struct f2fs_stat_info), GFP_KERNEL);
 	if (!si)
 		return -ENOMEM;
 

fs/f2fs/dir.c

@@ -713,6 +713,8 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
 
 	f2fs_update_time(F2FS_I_SB(dir), REQ_TIME);
 
+	add_ino_entry(F2FS_I_SB(dir), dir->i_ino, TRANS_DIR_INO);
+
 	if (f2fs_has_inline_dentry(dir))
 		return f2fs_delete_inline_entry(dentry, page, dir, inode);
 
@@ -798,6 +800,7 @@ int f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 	unsigned int bit_pos;
 	struct f2fs_dir_entry *de = NULL;
 	struct fscrypt_str de_name = FSTR_INIT(NULL, 0);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(d->inode);
 
 	bit_pos = ((unsigned long)ctx->pos % d->max);
 
@@ -836,6 +839,9 @@ int f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 					le32_to_cpu(de->ino), d_type))
 			return 1;
 
+		if (sbi->readdir_ra == 1)
+			ra_node_page(sbi, le32_to_cpu(de->ino));
+
 		bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len));
 		ctx->pos = start_pos + bit_pos;
 	}

fs/f2fs/f2fs.h

@@ -19,6 +19,7 @@
 #include <linux/magic.h>
 #include <linux/kobject.h>
 #include <linux/sched.h>
+#include <linux/cred.h>
 #include <linux/vmalloc.h>
 #include <linux/bio.h>
 #include <linux/blkdev.h>
@@ -43,6 +44,7 @@
 #ifdef CONFIG_F2FS_FAULT_INJECTION
 enum {
 	FAULT_KMALLOC,
+	FAULT_KVMALLOC,
 	FAULT_PAGE_ALLOC,
 	FAULT_PAGE_GET,
 	FAULT_ALLOC_BIO,
@@ -94,6 +96,7 @@ extern char *fault_name[FAULT_MAX];
 #define F2FS_MOUNT_PRJQUOTA		0x00200000
 #define F2FS_MOUNT_QUOTA		0x00400000
 #define F2FS_MOUNT_INLINE_XATTR_SIZE	0x00800000
+#define F2FS_MOUNT_RESERVE_ROOT		0x01000000
 
 #define clear_opt(sbi, option)	((sbi)->mount_opt.opt &= ~F2FS_MOUNT_##option)
 #define set_opt(sbi, option)	((sbi)->mount_opt.opt |= F2FS_MOUNT_##option)
@@ -121,6 +124,7 @@ struct f2fs_mount_info {
 #define F2FS_FEATURE_INODE_CHKSUM	0x0020
 #define F2FS_FEATURE_FLEXIBLE_INLINE_XATTR	0x0040
 #define F2FS_FEATURE_QUOTA_INO		0x0080
+#define F2FS_FEATURE_INODE_CRTIME	0x0100
 
 #define F2FS_HAS_FEATURE(sb, mask)					\
 	((F2FS_SB(sb)->raw_super->feature & cpu_to_le32(mask)) != 0)
@@ -129,6 +133,12 @@ struct f2fs_mount_info {
 #define F2FS_CLEAR_FEATURE(sb, mask)					\
 	(F2FS_SB(sb)->raw_super->feature &= ~cpu_to_le32(mask))
 
+/*
+ * Default values for user and/or group using reserved blocks
+ */
+#define	F2FS_DEF_RESUID		0
+#define	F2FS_DEF_RESGID		0
+
 /*
  * For checkpoint manager
  */
@@ -179,6 +189,7 @@ enum {
 	ORPHAN_INO,		/* for orphan ino list */
 	APPEND_INO,		/* for append ino list */
 	UPDATE_INO,		/* for update ino list */
+	TRANS_DIR_INO,		/* for trasactions dir ino list */
 	FLUSH_INO,		/* for multiple device flushing */
 	MAX_INO_ENTRY,		/* max. list */
 };
@@ -264,7 +275,6 @@ struct discard_cmd_control {
 	struct task_struct *f2fs_issue_discard;	/* discard thread */
 	struct list_head entry_list;		/* 4KB discard entry list */
 	struct list_head pend_list[MAX_PLIST_NUM];/* store pending entries */
-	unsigned char pend_list_tag[MAX_PLIST_NUM];/* tag for pending entries */
 	struct list_head wait_list;		/* store on-flushing entries */
 	struct list_head fstrim_list;		/* in-flight discard from fstrim */
 	wait_queue_head_t discard_wait_queue;	/* waiting queue for wake-up */
@@ -347,6 +357,9 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal,
 #define F2FS_IOC_GARBAGE_COLLECT_RANGE	_IOW(F2FS_IOCTL_MAGIC, 11,	\
 						struct f2fs_gc_range)
 #define F2FS_IOC_GET_FEATURES		_IOR(F2FS_IOCTL_MAGIC, 12, __u32)
+#define F2FS_IOC_SET_PIN_FILE		_IOW(F2FS_IOCTL_MAGIC, 13, __u32)
+#define F2FS_IOC_GET_PIN_FILE		_IOR(F2FS_IOCTL_MAGIC, 14, __u32)
+#define F2FS_IOC_PRECACHE_EXTENTS	_IO(F2FS_IOCTL_MAGIC, 15)
 
 #define F2FS_IOC_SET_ENCRYPTION_POLICY	FS_IOC_SET_ENCRYPTION_POLICY
 #define F2FS_IOC_GET_ENCRYPTION_POLICY	FS_IOC_GET_ENCRYPTION_POLICY
@@ -402,10 +415,9 @@ struct f2fs_flush_device {
 #define DEF_MIN_INLINE_SIZE		1
 static inline int get_extra_isize(struct inode *inode);
 static inline int get_inline_xattr_addrs(struct inode *inode);
-#define F2FS_INLINE_XATTR_ADDRS(inode)	get_inline_xattr_addrs(inode)
 #define MAX_INLINE_DATA(inode)	(sizeof(__le32) *			\
 				(CUR_ADDRS_PER_INODE(inode) -		\
-				F2FS_INLINE_XATTR_ADDRS(inode) -	\
+				get_inline_xattr_addrs(inode) -	\
 				DEF_INLINE_RESERVED_SIZE))
 
 /* for inline dir */
@@ -542,6 +554,8 @@ struct f2fs_map_blocks {
 	unsigned int m_len;
 	unsigned int m_flags;
 	pgoff_t *m_next_pgofs;		/* point next possible non-hole pgofs */
+	pgoff_t *m_next_extent;		/* point to next possible extent */
+	int m_seg_type;
 };
 
 /* for flag in get_data_block */
@@ -551,6 +565,7 @@ enum {
 	F2FS_GET_BLOCK_BMAP,
 	F2FS_GET_BLOCK_PRE_DIO,
 	F2FS_GET_BLOCK_PRE_AIO,
+	F2FS_GET_BLOCK_PRECACHE,
 };
 
 /*
@@ -583,7 +598,10 @@ struct f2fs_inode_info {
 	unsigned long i_flags;		/* keep an inode flags for ioctl */
 	unsigned char i_advise;		/* use to give file attribute hints */
 	unsigned char i_dir_level;	/* use for dentry level for large dir */
-	unsigned int i_current_depth;	/* use only in directory structure */
+	union {
+		unsigned int i_current_depth;	/* only for directory depth */
+		unsigned short i_gc_failures;	/* only for regular file */
+	};
 	unsigned int i_pino;		/* parent inode number */
 	umode_t i_acl_mode;		/* keep file acl mode temporarily */
 
@@ -618,6 +636,7 @@ struct f2fs_inode_info {
 	int i_extra_isize;		/* size of extra space located in i_addr */
 	kprojid_t i_projid;		/* id for project quota */
 	int i_inline_xattr_size;	/* inline xattr size */
+	struct timespec i_crtime;	/* inode creation time */
 };
 
 static inline void get_extent_info(struct extent_info *ext,
@@ -922,6 +941,7 @@ enum cp_reason_type {
 	CP_NODE_NEED_CP,
 	CP_FASTBOOT_MODE,
 	CP_SPEC_LOG_NUM,
+	CP_RECOVER_DIR,
 };
 
 enum iostat_type {
@@ -957,6 +977,7 @@ struct f2fs_io_info {
 	int need_lock;		/* indicate we need to lock cp_rwsem */
 	bool in_list;		/* indicate fio is in io_list */
 	enum iostat_type io_type;	/* io type */
+	struct writeback_control *io_wbc; /* writeback control */
 };
 
 #define is_read_io(rw) ((rw) == READ)
@@ -1093,6 +1114,7 @@ struct f2fs_sb_info {
 	int dir_level;				/* directory level */
 	int inline_xattr_size;			/* inline xattr size */
 	unsigned int trigger_ssr_threshold;	/* threshold to trigger ssr */
+	int readdir_ra;				/* readahead inode in readdir */
 
 	block_t user_block_count;		/* # of user blocks */
 	block_t total_valid_block_count;	/* # of valid blocks */
@@ -1100,6 +1122,11 @@ struct f2fs_sb_info {
 	block_t last_valid_block_count;		/* for recovery */
 	block_t reserved_blocks;		/* configurable reserved blocks */
 	block_t current_reserved_blocks;	/* current reserved blocks */
+	block_t root_reserved_blocks;		/* root reserved blocks */
+	kuid_t s_resuid;			/* reserved blocks for uid */
+	kgid_t s_resgid;			/* reserved blocks for gid */
+
+	unsigned int nquota_files;		/* # of quota sysfile */
 
 	u32 s_next_generation;			/* for NFS support */
 
@@ -1124,6 +1151,9 @@ struct f2fs_sb_info {
 	/* threshold for converting bg victims for fg */
 	u64 fggc_threshold;
 
+	/* threshold for gc trials on pinned files */
+	u64 gc_pin_file_threshold;
+
 	/* maximum # of trials to find a victim segment for SSR and GC */
 	unsigned int max_victim_search;
 
@@ -1250,33 +1280,7 @@ static inline bool is_idle(struct f2fs_sb_info *sbi)
 /*
  * Inline functions
  */
-static inline u32 f2fs_crc32(struct f2fs_sb_info *sbi, const void *address,
-			   unsigned int length)
-{
-	SHASH_DESC_ON_STACK(shash, sbi->s_chksum_driver);
-	u32 *ctx = (u32 *)shash_desc_ctx(shash);
-	u32 retval;
-	int err;
-
-	shash->tfm = sbi->s_chksum_driver;
-	shash->flags = 0;
-	*ctx = F2FS_SUPER_MAGIC;
-
-	err = crypto_shash_update(shash, address, length);
-	BUG_ON(err);
-
-	retval = *ctx;
-	barrier_data(ctx);
-	return retval;
-}
-
-static inline bool f2fs_crc_valid(struct f2fs_sb_info *sbi, __u32 blk_crc,
-				  void *buf, size_t buf_size)
-{
-	return f2fs_crc32(sbi, buf, buf_size) == blk_crc;
-}
-
-static inline u32 f2fs_chksum(struct f2fs_sb_info *sbi, u32 crc,
+static inline u32 __f2fs_crc32(struct f2fs_sb_info *sbi, u32 crc,
 			      const void *address, unsigned int length)
 {
 	struct {
@@ -1297,6 +1301,24 @@ static inline u32 f2fs_chksum(struct f2fs_sb_info *sbi, u32 crc,
 	return *(u32 *)desc.ctx;
 }
 
+static inline u32 f2fs_crc32(struct f2fs_sb_info *sbi, const void *address,
+			   unsigned int length)
+{
+	return __f2fs_crc32(sbi, F2FS_SUPER_MAGIC, address, length);
+}
+
+static inline bool f2fs_crc_valid(struct f2fs_sb_info *sbi, __u32 blk_crc,
+				  void *buf, size_t buf_size)
+{
+	return f2fs_crc32(sbi, buf, buf_size) == blk_crc;
+}
+
+static inline u32 f2fs_chksum(struct f2fs_sb_info *sbi, u32 crc,
+			      const void *address, unsigned int length)
+{
+	return __f2fs_crc32(sbi, crc, address, length);
+}
+
 static inline struct f2fs_inode_info *F2FS_I(struct inode *inode)
 {
 	return container_of(inode, struct f2fs_inode_info, vfs_inode);
@@ -1555,6 +1577,25 @@ static inline bool f2fs_has_xattr_block(unsigned int ofs)
 	return ofs == XATTR_NODE_OFFSET;
 }
 
+static inline bool __allow_reserved_blocks(struct f2fs_sb_info *sbi,
+					struct inode *inode)
+{
+	if (!inode)
+		return true;
+	if (!test_opt(sbi, RESERVE_ROOT))
+		return false;
+	if (IS_NOQUOTA(inode))
+		return true;
+	if (capable(CAP_SYS_RESOURCE))
+		return true;
+	if (uid_eq(sbi->s_resuid, current_fsuid()))
+		return true;
+	if (!gid_eq(sbi->s_resgid, GLOBAL_ROOT_GID) &&
+					in_group_p(sbi->s_resgid))
+		return true;
+	return false;
+}
+
 static inline void f2fs_i_blocks_write(struct inode *, block_t, bool, bool);
 static inline int inc_valid_block_count(struct f2fs_sb_info *sbi,
 				 struct inode *inode, blkcnt_t *count)
@@ -1584,11 +1625,17 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi,
 	sbi->total_valid_block_count += (block_t)(*count);
 	avail_user_block_count = sbi->user_block_count -
 					sbi->current_reserved_blocks;
+
+	if (!__allow_reserved_blocks(sbi, inode))
+		avail_user_block_count -= sbi->root_reserved_blocks;
+
 	if (unlikely(sbi->total_valid_block_count > avail_user_block_count)) {
 		diff = sbi->total_valid_block_count - avail_user_block_count;
+		if (diff > *count)
+			diff = *count;
 		*count -= diff;
 		release = diff;
-		sbi->total_valid_block_count = avail_user_block_count;
+		sbi->total_valid_block_count -= diff;
 		if (!*count) {
 			spin_unlock(&sbi->stat_lock);
 			percpu_counter_sub(&sbi->alloc_valid_block_count, diff);
@@ -1597,7 +1644,7 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi,
 	}
 	spin_unlock(&sbi->stat_lock);
 
-	if (release)
+	if (unlikely(release))
 		dquot_release_reservation_block(inode, release);
 	f2fs_i_blocks_write(inode, *count, true, true);
 	return 0;
@@ -1777,9 +1824,13 @@ static inline int inc_valid_node_count(struct f2fs_sb_info *sbi,
 
 	spin_lock(&sbi->stat_lock);
 
-	valid_block_count = sbi->total_valid_block_count + 1;
-	if (unlikely(valid_block_count + sbi->current_reserved_blocks >
-						sbi->user_block_count)) {
+	valid_block_count = sbi->total_valid_block_count +
+					sbi->current_reserved_blocks + 1;
+
+	if (!__allow_reserved_blocks(sbi, inode))
+		valid_block_count += sbi->root_reserved_blocks;
+
+	if (unlikely(valid_block_count > sbi->user_block_count)) {
 		spin_unlock(&sbi->stat_lock);
 		goto enospc;
 	}
@@ -1992,11 +2043,11 @@ static inline block_t datablock_addr(struct inode *inode,
 	raw_node = F2FS_NODE(node_page);
 
 	/* from GC path only */
-	if (!inode) {
-		if (is_inode)
+	if (is_inode) {
+		if (!inode)
 			base = offset_in_addr(&raw_node->i);
-	} else if (f2fs_has_extra_attr(inode) && is_inode) {
-		base = get_extra_isize(inode);
+		else if (f2fs_has_extra_attr(inode))
+			base = get_extra_isize(inode);
 	}
 
 	addr_array = blkaddr_in_node(raw_node);
@@ -2107,6 +2158,7 @@ enum {
 	FI_HOT_DATA,		/* indicate file is hot */
 	FI_EXTRA_ATTR,		/* indicate file has extra attribute */
 	FI_PROJ_INHERIT,	/* indicate file inherits projectid */
+	FI_PIN_FILE,		/* indicate file should not be gced */
 };
 
 static inline void __mark_inode_dirty_flag(struct inode *inode,
@@ -2116,10 +2168,12 @@ static inline void __mark_inode_dirty_flag(struct inode *inode,
 	case FI_INLINE_XATTR:
 	case FI_INLINE_DATA:
 	case FI_INLINE_DENTRY:
+	case FI_NEW_INODE:
 		if (set)
 			return;
 	case FI_DATA_EXIST:
 	case FI_INLINE_DOTS:
+	case FI_PIN_FILE:
 		f2fs_mark_inode_dirty_sync(inode, true);
 	}
 }
@@ -2200,6 +2254,13 @@ static inline void f2fs_i_depth_write(struct inode *inode, unsigned int depth)
 	f2fs_mark_inode_dirty_sync(inode, true);
 }
 
+static inline void f2fs_i_gc_failures_write(struct inode *inode,
+					unsigned int count)
+{
+	F2FS_I(inode)->i_gc_failures = count;
+	f2fs_mark_inode_dirty_sync(inode, true);
+}
+
 static inline void f2fs_i_xnid_write(struct inode *inode, nid_t xnid)
 {
 	F2FS_I(inode)->i_xattr_nid = xnid;
@@ -2228,6 +2289,8 @@ static inline void get_inline_info(struct inode *inode, struct f2fs_inode *ri)
 		set_bit(FI_INLINE_DOTS, &fi->flags);
 	if (ri->i_inline & F2FS_EXTRA_ATTR)
 		set_bit(FI_EXTRA_ATTR, &fi->flags);
+	if (ri->i_inline & F2FS_PIN_FILE)
+		set_bit(FI_PIN_FILE, &fi->flags);
 }
 
 static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri)
@@ -2246,6 +2309,8 @@ static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri)
 		ri->i_inline |= F2FS_INLINE_DOTS;
 	if (is_inode_flag_set(inode, FI_EXTRA_ATTR))
 		ri->i_inline |= F2FS_EXTRA_ATTR;
+	if (is_inode_flag_set(inode, FI_PIN_FILE))
+		ri->i_inline |= F2FS_PIN_FILE;
 }
 
 static inline int f2fs_has_extra_attr(struct inode *inode)
@@ -2260,7 +2325,7 @@ static inline int f2fs_has_inline_xattr(struct inode *inode)
 
 static inline unsigned int addrs_per_inode(struct inode *inode)
 {
-	return CUR_ADDRS_PER_INODE(inode) - F2FS_INLINE_XATTR_ADDRS(inode);
+	return CUR_ADDRS_PER_INODE(inode) - get_inline_xattr_addrs(inode);
 }
 
 static inline void *inline_xattr_addr(struct inode *inode, struct page *page)
@@ -2268,7 +2333,7 @@ static inline void *inline_xattr_addr(struct inode *inode, struct page *page)
 	struct f2fs_inode *ri = F2FS_INODE(page);
 
 	return (void *)&(ri->i_addr[DEF_ADDRS_PER_INODE -
-					F2FS_INLINE_XATTR_ADDRS(inode)]);
+					get_inline_xattr_addrs(inode)]);
 }
 
 static inline int inline_xattr_size(struct inode *inode)
@@ -2291,6 +2356,11 @@ static inline int f2fs_has_inline_dots(struct inode *inode)
 	return is_inode_flag_set(inode, FI_INLINE_DOTS);
 }
 
+static inline bool f2fs_is_pinned_file(struct inode *inode)
+{
+	return is_inode_flag_set(inode, FI_PIN_FILE);
+}
+
 static inline bool f2fs_is_atomic_file(struct inode *inode)
 {
 	return is_inode_flag_set(inode, FI_ATOMIC_FILE);
@@ -2418,12 +2488,35 @@ static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi,
 	return kmalloc(size, flags);
 }
 
+static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi,
+					size_t size, gfp_t flags)
+{
+	return f2fs_kmalloc(sbi, size, flags | __GFP_ZERO);
+}
+
+static inline void *f2fs_kvmalloc(struct f2fs_sb_info *sbi,
+					size_t size, gfp_t flags)
+{
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_KVMALLOC)) {
+		f2fs_show_injection_info(FAULT_KVMALLOC);
+		return NULL;
+	}
+#endif
+	return kvmalloc(size, flags);
+}
+
+static inline void *f2fs_kvzalloc(struct f2fs_sb_info *sbi,
+					size_t size, gfp_t flags)
+{
+	return f2fs_kvmalloc(sbi, size, flags | __GFP_ZERO);
+}
+
 static inline int get_extra_isize(struct inode *inode)
 {
 	return F2FS_I(inode)->i_extra_isize / sizeof(__le32);
 }
 
-static inline int f2fs_sb_has_flexible_inline_xattr(struct super_block *sb);
 static inline int get_inline_xattr_addrs(struct inode *inode)
 {
 	return F2FS_I(inode)->i_inline_xattr_size;
@@ -2479,9 +2572,11 @@ int f2fs_getattr(const struct path *path, struct kstat *stat,
 			u32 request_mask, unsigned int flags);
 int f2fs_setattr(struct dentry *dentry, struct iattr *attr);
 int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end);
-int truncate_data_blocks_range(struct dnode_of_data *dn, int count);
+void truncate_data_blocks_range(struct dnode_of_data *dn, int count);
+int f2fs_precache_extents(struct inode *inode);
 long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg);
 long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
+int f2fs_pin_file_control(struct inode *inode, bool inc);
 
 /*
  * inode.c
@@ -2492,8 +2587,8 @@ void f2fs_inode_chksum_set(struct f2fs_sb_info *sbi, struct page *page);
 struct inode *f2fs_iget(struct super_block *sb, unsigned long ino);
 struct inode *f2fs_iget_retry(struct super_block *sb, unsigned long ino);
 int try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink);
-int update_inode(struct inode *inode, struct page *node_page);
-int update_inode_page(struct inode *inode);
+void update_inode(struct inode *inode, struct page *node_page);
+void update_inode_page(struct inode *inode);
 int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc);
 void f2fs_evict_inode(struct inode *inode);
 void handle_failed_inode(struct inode *inode);
@@ -2604,10 +2699,9 @@ void alloc_nid_done(struct f2fs_sb_info *sbi, nid_t nid);
 void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid);
 int try_to_free_nids(struct f2fs_sb_info *sbi, int nr_shrink);
 void recover_inline_xattr(struct inode *inode, struct page *page);
-int recover_xattr_data(struct inode *inode, struct page *page,
-			block_t blkaddr);
+int recover_xattr_data(struct inode *inode, struct page *page);
 int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page);
-int restore_node_summary(struct f2fs_sb_info *sbi,
+void restore_node_summary(struct f2fs_sb_info *sbi,
 			unsigned int segno, struct f2fs_summary_block *sum);
 void flush_nat_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc);
 int build_node_manager(struct f2fs_sb_info *sbi);
@@ -2634,6 +2728,7 @@ void invalidate_blocks(struct f2fs_sb_info *sbi, block_t addr);
 bool is_checkpointed_data(struct f2fs_sb_info *sbi, block_t blkaddr);
 void init_discard_policy(struct discard_policy *dpolicy, int discard_type,
 						unsigned int granularity);
+void drop_discard_cmd(struct f2fs_sb_info *sbi);
 void stop_discard_thread(struct f2fs_sb_info *sbi);
 bool f2fs_wait_discard_bios(struct f2fs_sb_info *sbi);
 void clear_prefree_segments(struct f2fs_sb_info *sbi, struct cp_control *cpc);
@@ -2672,6 +2767,7 @@ int build_segment_manager(struct f2fs_sb_info *sbi);
 void destroy_segment_manager(struct f2fs_sb_info *sbi);
 int __init create_segment_manager_caches(void);
 void destroy_segment_manager_caches(void);
+int rw_hint_to_seg_type(enum rw_hint hint);
 
 /*
  * checkpoint.c
@@ -2741,6 +2837,8 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
 			int create, int flag);
 int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			u64 start, u64 len);
+bool should_update_inplace(struct inode *inode, struct f2fs_io_info *fio);
+bool should_update_outplace(struct inode *inode, struct f2fs_io_info *fio);
 void f2fs_set_page_dirty_nobuffers(struct page *page);
 int __f2fs_write_data_pages(struct address_space *mapping,
 						struct writeback_control *wbc,
@@ -3109,6 +3207,11 @@ static inline int f2fs_sb_has_quota_ino(struct super_block *sb)
 	return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_QUOTA_INO);
 }
 
+static inline int f2fs_sb_has_inode_crtime(struct super_block *sb)
+{
+	return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_INODE_CRTIME);
+}
+
 #ifdef CONFIG_BLK_DEV_ZONED
 static inline int get_blkz_type(struct f2fs_sb_info *sbi,
 			struct block_device *bdev, block_t blkaddr)

fs/f2fs/file.c

@@ -165,6 +165,9 @@ static inline enum cp_reason_type need_do_checkpoint(struct inode *inode)
 		cp_reason = CP_FASTBOOT_MODE;
 	else if (sbi->active_logs == 2)
 		cp_reason = CP_SPEC_LOG_NUM;
+	else if (need_dentry_mark(sbi, inode->i_ino) &&
+		exist_written_data(sbi, F2FS_I(inode)->i_pino, TRANS_DIR_INO))
+		cp_reason = CP_RECOVER_DIR;
 
 	return cp_reason;
 }
@@ -472,26 +475,14 @@ static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma)
 
 static int f2fs_file_open(struct inode *inode, struct file *filp)
 {
-	struct dentry *dir;
+	int err = fscrypt_file_open(inode, filp);
 
-	if (f2fs_encrypted_inode(inode)) {
-		int ret = fscrypt_get_encryption_info(inode);
-		if (ret)
-			return -EACCES;
-		if (!fscrypt_has_encryption_key(inode))
-			return -ENOKEY;
-	}
-	dir = dget_parent(file_dentry(filp));
-	if (f2fs_encrypted_inode(d_inode(dir)) &&
-			!fscrypt_has_permitted_context(d_inode(dir), inode)) {
-		dput(dir);
-		return -EPERM;
-	}
-	dput(dir);
+	if (err)
+		return err;
 	return dquot_file_open(inode, filp);
 }
 
-int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
+void truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
 	struct f2fs_node *raw_node;
@@ -534,7 +525,6 @@ int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 	f2fs_update_time(sbi, REQ_TIME);
 	trace_f2fs_truncate_data_blocks_range(dn->inode, dn->nid,
 					 dn->ofs_in_node, nr_free);
-	return nr_free;
 }
 
 void truncate_data_blocks(struct dnode_of_data *dn)
@@ -682,8 +672,17 @@ int f2fs_getattr(const struct path *path, struct kstat *stat,
 {
 	struct inode *inode = d_inode(path->dentry);
 	struct f2fs_inode_info *fi = F2FS_I(inode);
+	struct f2fs_inode *ri;
 	unsigned int flags;
 
+	if (f2fs_has_extra_attr(inode) &&
+			f2fs_sb_has_inode_crtime(inode->i_sb) &&
+			F2FS_FITS_IN_INODE(ri, fi->i_extra_isize, i_crtime)) {
+		stat->result_mask |= STATX_BTIME;
+		stat->btime.tv_sec = fi->i_crtime.tv_sec;
+		stat->btime.tv_nsec = fi->i_crtime.tv_nsec;
+	}
+
 	flags = fi->i_flags & (FS_FL_USER_VISIBLE | FS_PROJINHERIT_FL);
 	if (flags & FS_APPEND_FL)
 		stat->attributes |= STATX_ATTR_APPEND;
@@ -755,6 +754,10 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
 	if (err)
 		return err;
 
+	err = fscrypt_prepare_setattr(dentry, attr);
+	if (err)
+		return err;
+
 	if (is_quota_modification(inode, attr)) {
 		err = dquot_initialize(inode);
 		if (err)
@@ -770,14 +773,6 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
 	}
 
 	if (attr->ia_valid & ATTR_SIZE) {
-		if (f2fs_encrypted_inode(inode)) {
-			err = fscrypt_get_encryption_info(inode);
-			if (err)
-				return err;
-			if (!fscrypt_has_encryption_key(inode))
-				return -ENOKEY;
-		}
-
 		if (attr->ia_size <= i_size_read(inode)) {
 			down_write(&F2FS_I(inode)->i_mmap_sem);
 			truncate_setsize(inode, attr->ia_size);
@@ -1114,11 +1109,13 @@ static int __exchange_data_block(struct inode *src_inode,
 	while (len) {
 		olen = min((pgoff_t)4 * ADDRS_PER_BLOCK, len);
 
-		src_blkaddr = kvzalloc(sizeof(block_t) * olen, GFP_KERNEL);
+		src_blkaddr = f2fs_kvzalloc(F2FS_I_SB(src_inode),
+					sizeof(block_t) * olen, GFP_KERNEL);
 		if (!src_blkaddr)
 			return -ENOMEM;
 
-		do_replace = kvzalloc(sizeof(int) * olen, GFP_KERNEL);
+		do_replace = f2fs_kvzalloc(F2FS_I_SB(src_inode),
+					sizeof(int) * olen, GFP_KERNEL);
 		if (!do_replace) {
 			kvfree(src_blkaddr);
 			return -ENOMEM;
@@ -1186,14 +1183,14 @@ static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	pg_start = offset >> PAGE_SHIFT;
 	pg_end = (offset + len) >> PAGE_SHIFT;
 
+	/* avoid gc operation during block exchange */
+	down_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
+
 	down_write(&F2FS_I(inode)->i_mmap_sem);
 	/* write out all dirty pages from offset */
 	ret = filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
 	if (ret)
-		goto out;
-
-	/* avoid gc operation during block exchange */
-	down_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
+		goto out_unlock;
 
 	truncate_pagecache(inode, offset);
 
@@ -1212,9 +1209,8 @@ static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	if (!ret)
 		f2fs_i_size_write(inode, new_size);
 out_unlock:
-	up_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
-out:
 	up_write(&F2FS_I(inode)->i_mmap_sem);
+	up_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
 	return ret;
 }
 
@@ -1385,6 +1381,9 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len)
 
 	f2fs_balance_fs(sbi, true);
 
+	/* avoid gc operation during block exchange */
+	down_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
+
 	down_write(&F2FS_I(inode)->i_mmap_sem);
 	ret = truncate_blocks(inode, i_size_read(inode), true);
 	if (ret)
@@ -1395,9 +1394,6 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len)
 	if (ret)
 		goto out;
 
-	/* avoid gc operation during block exchange */
-	down_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
-
 	truncate_pagecache(inode, offset);
 
 	pg_start = offset >> PAGE_SHIFT;
@@ -1425,10 +1421,9 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len)
 
 	if (!ret)
 		f2fs_i_size_write(inode, new_size);
-
-	up_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
 out:
 	up_write(&F2FS_I(inode)->i_mmap_sem);
+	up_write(&F2FS_I(inode)->dio_rwsem[WRITE]);
 	return ret;
 }
 
@@ -1436,7 +1431,8 @@ static int expand_inode_data(struct inode *inode, loff_t offset,
 					loff_t len, int mode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct f2fs_map_blocks map = { .m_next_pgofs = NULL };
+	struct f2fs_map_blocks map = { .m_next_pgofs = NULL,
+			.m_next_extent = NULL, .m_seg_type = NO_CHECK_TYPE };
 	pgoff_t pg_end;
 	loff_t new_size = i_size_read(inode);
 	loff_t off_end;
@@ -1852,14 +1848,20 @@ static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg)
 	switch (in) {
 	case F2FS_GOING_DOWN_FULLSYNC:
 		sb = freeze_bdev(sb->s_bdev);
-		if (sb && !IS_ERR(sb)) {
+		if (IS_ERR(sb)) {
+			ret = PTR_ERR(sb);
+			goto out;
+		}
+		if (sb) {
 			f2fs_stop_checkpoint(sbi, false);
 			thaw_bdev(sb->s_bdev, sb);
 		}
 		break;
 	case F2FS_GOING_DOWN_METASYNC:
 		/* do checkpoint only */
-		f2fs_sync_fs(sb, 1);
+		ret = f2fs_sync_fs(sb, 1);
+		if (ret)
+			goto out;
 		f2fs_stop_checkpoint(sbi, false);
 		break;
 	case F2FS_GOING_DOWN_NOSYNC:
@@ -1873,6 +1875,13 @@ static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg)
 		ret = -EINVAL;
 		goto out;
 	}
+
+	stop_gc_thread(sbi);
+	stop_discard_thread(sbi);
+
+	drop_discard_cmd(sbi);
+	clear_opt(sbi, DISCARD);
+
 	f2fs_update_time(sbi, REQ_TIME);
 out:
 	mnt_drop_write_file(filp);
@@ -2084,9 +2093,10 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
 					struct f2fs_defragment *range)
 {
 	struct inode *inode = file_inode(filp);
-	struct f2fs_map_blocks map = { .m_next_pgofs = NULL };
+	struct f2fs_map_blocks map = { .m_next_extent = NULL,
+					.m_seg_type = NO_CHECK_TYPE };
 	struct extent_info ei = {0,0,0};
-	pgoff_t pg_start, pg_end;
+	pgoff_t pg_start, pg_end, next_pgofs;
 	unsigned int blk_per_seg = sbi->blocks_per_seg;
 	unsigned int total = 0, sec_num;
 	block_t blk_end = 0;
@@ -2094,7 +2104,7 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
 	int err;
 
 	/* if in-place-update policy is enabled, don't waste time here */
-	if (need_inplace_update_policy(inode, NULL))
+	if (should_update_inplace(inode, NULL))
 		return -EINVAL;
 
 	pg_start = range->start >> PAGE_SHIFT;
@@ -2120,6 +2130,7 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
 	}
 
 	map.m_lblk = pg_start;
+	map.m_next_pgofs = &next_pgofs;
 
 	/*
 	 * lookup mapping info in dnode page cache, skip defragmenting if all
@@ -2133,14 +2144,16 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
 			goto out;
 
 		if (!(map.m_flags & F2FS_MAP_FLAGS)) {
-			map.m_lblk++;
+			map.m_lblk = next_pgofs;
 			continue;
 		}
 
-		if (blk_end && blk_end != map.m_pblk) {
+		if (blk_end && blk_end != map.m_pblk)
 			fragmented = true;
-			break;
-		}
+
+		/* record total count of block that we're going to move */
+		total += map.m_len;
+
 		blk_end = map.m_pblk + map.m_len;
 
 		map.m_lblk += map.m_len;
@@ -2149,10 +2162,7 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
 	if (!fragmented)
 		goto out;
 
-	map.m_lblk = pg_start;
-	map.m_len = pg_end - pg_start;
-
-	sec_num = (map.m_len + BLKS_PER_SEC(sbi) - 1) / BLKS_PER_SEC(sbi);
+	sec_num = (total + BLKS_PER_SEC(sbi) - 1) / BLKS_PER_SEC(sbi);
 
 	/*
 	 * make sure there are enough free section for LFS allocation, this can
@@ -2164,6 +2174,10 @@ static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
 		goto out;
 	}
 
+	map.m_lblk = pg_start;
+	map.m_len = pg_end - pg_start;
+	total = 0;
+
 	while (map.m_lblk < pg_end) {
 		pgoff_t idx;
 		int cnt = 0;
@@ -2175,7 +2189,7 @@ do_map:
 			goto clear_out;
 
 		if (!(map.m_flags & F2FS_MAP_FLAGS)) {
-			map.m_lblk++;
+			map.m_lblk = next_pgofs;
 			continue;
 		}
 
@@ -2681,6 +2695,125 @@ static int f2fs_ioc_fssetxattr(struct file *filp, unsigned long arg)
 	return 0;
 }
 
+int f2fs_pin_file_control(struct inode *inode, bool inc)
+{
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+
+	/* Use i_gc_failures for normal file as a risk signal. */
+	if (inc)
+		f2fs_i_gc_failures_write(inode, fi->i_gc_failures + 1);
+
+	if (fi->i_gc_failures > sbi->gc_pin_file_threshold) {
+		f2fs_msg(sbi->sb, KERN_WARNING,
+			"%s: Enable GC = ino %lx after %x GC trials\n",
+			__func__, inode->i_ino, fi->i_gc_failures);
+		clear_inode_flag(inode, FI_PIN_FILE);
+		return -EAGAIN;
+	}
+	return 0;
+}
+
+static int f2fs_ioc_set_pin_file(struct file *filp, unsigned long arg)
+{
+	struct inode *inode = file_inode(filp);
+	__u32 pin;
+	int ret = 0;
+
+	if (!inode_owner_or_capable(inode))
+		return -EACCES;
+
+	if (get_user(pin, (__u32 __user *)arg))
+		return -EFAULT;
+
+	if (!S_ISREG(inode->i_mode))
+		return -EINVAL;
+
+	if (f2fs_readonly(F2FS_I_SB(inode)->sb))
+		return -EROFS;
+
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	inode_lock(inode);
+
+	if (should_update_outplace(inode, NULL)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (!pin) {
+		clear_inode_flag(inode, FI_PIN_FILE);
+		F2FS_I(inode)->i_gc_failures = 1;
+		goto done;
+	}
+
+	if (f2fs_pin_file_control(inode, false)) {
+		ret = -EAGAIN;
+		goto out;
+	}
+	ret = f2fs_convert_inline_inode(inode);
+	if (ret)
+		goto out;
+
+	set_inode_flag(inode, FI_PIN_FILE);
+	ret = F2FS_I(inode)->i_gc_failures;
+done:
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
+out:
+	inode_unlock(inode);
+	mnt_drop_write_file(filp);
+	return ret;
+}
+
+static int f2fs_ioc_get_pin_file(struct file *filp, unsigned long arg)
+{
+	struct inode *inode = file_inode(filp);
+	__u32 pin = 0;
+
+	if (is_inode_flag_set(inode, FI_PIN_FILE))
+		pin = F2FS_I(inode)->i_gc_failures;
+	return put_user(pin, (u32 __user *)arg);
+}
+
+int f2fs_precache_extents(struct inode *inode)
+{
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+	struct f2fs_map_blocks map;
+	pgoff_t m_next_extent;
+	loff_t end;
+	int err;
+
+	if (is_inode_flag_set(inode, FI_NO_EXTENT))
+		return -EOPNOTSUPP;
+
+	map.m_lblk = 0;
+	map.m_next_pgofs = NULL;
+	map.m_next_extent = &m_next_extent;
+	map.m_seg_type = NO_CHECK_TYPE;
+	end = F2FS_I_SB(inode)->max_file_blocks;
+
+	while (map.m_lblk < end) {
+		map.m_len = end - map.m_lblk;
+
+		down_write(&fi->dio_rwsem[WRITE]);
+		err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_PRECACHE);
+		up_write(&fi->dio_rwsem[WRITE]);
+		if (err)
+			return err;
+
+		map.m_lblk = m_next_extent;
+	}
+
+	return err;
+}
+
+static int f2fs_ioc_precache_extents(struct file *filp, unsigned long arg)
+{
+	return f2fs_precache_extents(file_inode(filp));
+}
+
 long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	if (unlikely(f2fs_cp_error(F2FS_I_SB(file_inode(filp)))))
@@ -2731,6 +2864,12 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		return f2fs_ioc_fsgetxattr(filp, arg);
 	case F2FS_IOC_FSSETXATTR:
 		return f2fs_ioc_fssetxattr(filp, arg);
+	case F2FS_IOC_GET_PIN_FILE:
+		return f2fs_ioc_get_pin_file(filp, arg);
+	case F2FS_IOC_SET_PIN_FILE:
+		return f2fs_ioc_set_pin_file(filp, arg);
+	case F2FS_IOC_PRECACHE_EXTENTS:
+		return f2fs_ioc_precache_extents(filp, arg);
 	default:
 		return -ENOTTY;
 	}
@@ -2806,6 +2945,9 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case F2FS_IOC_GET_FEATURES:
 	case F2FS_IOC_FSGETXATTR:
 	case F2FS_IOC_FSSETXATTR:
+	case F2FS_IOC_GET_PIN_FILE:
+	case F2FS_IOC_SET_PIN_FILE:
+	case F2FS_IOC_PRECACHE_EXTENTS:
 		break;
 	default:
 		return -ENOIOCTLCMD;

fs/f2fs/gc.c

@@ -624,6 +624,11 @@ static void move_data_block(struct inode *inode, block_t bidx,
 	if (f2fs_is_atomic_file(inode))
 		goto out;
 
+	if (f2fs_is_pinned_file(inode)) {
+		f2fs_pin_file_control(inode, true);
+		goto out;
+	}
+
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = get_dnode_of_data(&dn, bidx, LOOKUP_NODE);
 	if (err)
@@ -686,7 +691,12 @@ static void move_data_block(struct inode *inode, block_t bidx,
 	fio.op = REQ_OP_WRITE;
 	fio.op_flags = REQ_SYNC;
 	fio.new_blkaddr = newaddr;
-	f2fs_submit_page_write(&fio);
+	err = f2fs_submit_page_write(&fio);
+	if (err) {
+		if (PageWriteback(fio.encrypted_page))
+			end_page_writeback(fio.encrypted_page);
+		goto put_page_out;
+	}
 
 	f2fs_update_iostat(fio.sbi, FS_GC_DATA_IO, F2FS_BLKSIZE);
 
@@ -720,6 +730,11 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type,
 
 	if (f2fs_is_atomic_file(inode))
 		goto out;
+	if (f2fs_is_pinned_file(inode)) {
+		if (gc_type == FG_GC)
+			f2fs_pin_file_control(inode, true);
+		goto out;
+	}
 
 	if (gc_type == BG_GC) {
 		if (PageWriteback(page))
@@ -1091,6 +1106,7 @@ void build_gc_manager(struct f2fs_sb_info *sbi)
 
 	sbi->fggc_threshold = div64_u64((main_count - ovp_count) *
 				BLKS_PER_SEC(sbi), (main_count - resv_count));
+	sbi->gc_pin_file_threshold = DEF_GC_FAILED_PINNED_FILES;
 
 	/* give warm/cold data area from slower device */
 	if (sbi->s_ndevs && sbi->segs_per_sec == 1)

fs/f2fs/gc.h

@@ -20,6 +20,8 @@
 #define LIMIT_INVALID_BLOCK	40 /* percentage over total user space */
 #define LIMIT_FREE_BLOCK	40 /* percentage over invalid + free space */
 
+#define DEF_GC_FAILED_PINNED_FILES	2048
+
 /* Search max. number of dirty segments to select a victim segment */
 #define DEF_MAX_VICTIM_SEARCH 4096 /* covers 8GB */
 

fs/f2fs/inode.c

@@ -22,6 +22,9 @@
 
 void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
 {
+	if (is_inode_flag_set(inode, FI_NEW_INODE))
+		return;
+
 	if (f2fs_inode_dirtied(inode, sync))
 		return;
 
@@ -275,6 +278,12 @@ static int do_read_inode(struct inode *inode)
 		i_projid = F2FS_DEF_PROJID;
 	fi->i_projid = make_kprojid(&init_user_ns, i_projid);
 
+	if (f2fs_has_extra_attr(inode) && f2fs_sb_has_inode_crtime(sbi->sb) &&
+			F2FS_FITS_IN_INODE(ri, fi->i_extra_isize, i_crtime)) {
+		fi->i_crtime.tv_sec = le64_to_cpu(ri->i_crtime);
+		fi->i_crtime.tv_nsec = le32_to_cpu(ri->i_crtime_nsec);
+	}
+
 	f2fs_put_page(node_page, 1);
 
 	stat_inc_inline_xattr(inode);
@@ -360,14 +369,15 @@ retry:
 	return inode;
 }
 
-int update_inode(struct inode *inode, struct page *node_page)
+void update_inode(struct inode *inode, struct page *node_page)
 {
 	struct f2fs_inode *ri;
 	struct extent_tree *et = F2FS_I(inode)->extent_tree;
 
-	f2fs_inode_synced(inode);
-
 	f2fs_wait_on_page_writeback(node_page, NODE, true);
+	set_page_dirty(node_page);
+
+	f2fs_inode_synced(inode);
 
 	ri = F2FS_INODE(node_page);
 
@@ -417,6 +427,15 @@ int update_inode(struct inode *inode, struct page *node_page)
 						F2FS_I(inode)->i_projid);
 			ri->i_projid = cpu_to_le32(i_projid);
 		}
+
+		if (f2fs_sb_has_inode_crtime(F2FS_I_SB(inode)->sb) &&
+			F2FS_FITS_IN_INODE(ri, F2FS_I(inode)->i_extra_isize,
+								i_crtime)) {
+			ri->i_crtime =
+				cpu_to_le64(F2FS_I(inode)->i_crtime.tv_sec);
+			ri->i_crtime_nsec =
+				cpu_to_le32(F2FS_I(inode)->i_crtime.tv_nsec);
+		}
 	}
 
 	__set_inode_rdev(inode, ri);
@@ -426,14 +445,12 @@ int update_inode(struct inode *inode, struct page *node_page)
 	if (inode->i_nlink == 0)
 		clear_inline_node(node_page);
 
-	return set_page_dirty(node_page);
 }
 
-int update_inode_page(struct inode *inode)
+void update_inode_page(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct page *node_page;
-	int ret = 0;
 retry:
 	node_page = get_node_page(sbi, inode->i_ino);
 	if (IS_ERR(node_page)) {
@@ -444,11 +461,10 @@ retry:
 		} else if (err != -ENOENT) {
 			f2fs_stop_checkpoint(sbi, false);
 		}
-		return 0;
+		return;
 	}
-	ret = update_inode(inode, node_page);
+	update_inode(inode, node_page);
 	f2fs_put_page(node_page, 1);
-	return ret;
 }
 
 int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)

fs/f2fs/namei.c

@@ -50,7 +50,8 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode)
 
 	inode->i_ino = ino;
 	inode->i_blocks = 0;
-	inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
+	inode->i_mtime = inode->i_atime = inode->i_ctime =
+			F2FS_I(inode)->i_crtime = current_time(inode);
 	inode->i_generation = sbi->s_next_generation++;
 
 	err = insert_inode_locked(inode);
@@ -74,12 +75,12 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode)
 	if (err)
 		goto fail_drop;
 
+	set_inode_flag(inode, FI_NEW_INODE);
+
 	/* If the directory encrypted, then we should encrypt the inode. */
 	if (f2fs_encrypted_inode(dir) && f2fs_may_encrypt(inode))
 		f2fs_set_encrypted_inode(inode);
 
-	set_inode_flag(inode, FI_NEW_INODE);
-
 	if (f2fs_sb_has_extra_attr(sbi->sb)) {
 		set_inode_flag(inode, FI_EXTRA_ATTR);
 		F2FS_I(inode)->i_extra_isize = F2FS_TOTAL_EXTRA_ATTR_SIZE;
@@ -240,9 +241,9 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
 
-	if (f2fs_encrypted_inode(dir) &&
-			!fscrypt_has_permitted_context(dir, inode))
-		return -EPERM;
+	err = fscrypt_prepare_link(old_dentry, dir, dentry);
+	if (err)
+		return err;
 
 	if (is_inode_flag_set(dir, FI_PROJ_INHERIT) &&
 			(!projid_eq(F2FS_I(dir)->i_projid,
@@ -357,20 +358,9 @@ static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry,
 
 	trace_f2fs_lookup_start(dir, dentry, flags);
 
-	if (f2fs_encrypted_inode(dir)) {
-		err = fscrypt_get_encryption_info(dir);
-
-		/*
-		 * DCACHE_ENCRYPTED_WITH_KEY is set if the dentry is
-		 * created while the directory was encrypted and we
-		 * don't have access to the key.
-		 */
-		if (fscrypt_has_encryption_key(dir))
-			fscrypt_set_encrypted_dentry(dentry);
-		fscrypt_set_d_op(dentry);
-		if (err && err != -ENOKEY)
-			goto out;
-	}
+	err = fscrypt_prepare_lookup(dir, dentry, flags);
+	if (err)
+		goto out;
 
 	if (dentry->d_name.len > F2FS_NAME_LEN) {
 		err = -ENAMETOOLONG;
@@ -544,7 +534,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 		struct qstr istr = QSTR_INIT(symname, len);
 		struct fscrypt_str ostr;
 
-		sd = kzalloc(disk_link.len, GFP_NOFS);
+		sd = f2fs_kzalloc(sbi, disk_link.len, GFP_NOFS);
 		if (!sd) {
 			err = -ENOMEM;
 			goto err_out;
@@ -800,18 +790,6 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
 
-	if ((f2fs_encrypted_inode(old_dir) &&
-			!fscrypt_has_encryption_key(old_dir)) ||
-			(f2fs_encrypted_inode(new_dir) &&
-			!fscrypt_has_encryption_key(new_dir)))
-		return -ENOKEY;
-
-	if ((old_dir != new_dir) && f2fs_encrypted_inode(new_dir) &&
-			!fscrypt_has_permitted_context(new_dir, old_inode)) {
-		err = -EPERM;
-		goto out;
-	}
-
 	if (is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
 			(!projid_eq(F2FS_I(new_dir)->i_projid,
 			F2FS_I(old_dentry->d_inode)->i_projid)))
@@ -958,6 +936,7 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		}
 		f2fs_i_links_write(old_dir, false);
 	}
+	add_ino_entry(sbi, new_dir->i_ino, TRANS_DIR_INO);
 
 	f2fs_unlock_op(sbi);
 
@@ -1002,18 +981,6 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
 
-	if ((f2fs_encrypted_inode(old_dir) &&
-			!fscrypt_has_encryption_key(old_dir)) ||
-			(f2fs_encrypted_inode(new_dir) &&
-			!fscrypt_has_encryption_key(new_dir)))
-		return -ENOKEY;
-
-	if ((f2fs_encrypted_inode(old_dir) || f2fs_encrypted_inode(new_dir)) &&
-			(old_dir != new_dir) &&
-			(!fscrypt_has_permitted_context(new_dir, old_inode) ||
-			 !fscrypt_has_permitted_context(old_dir, new_inode)))
-		return -EPERM;
-
 	if ((is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
 			!projid_eq(F2FS_I(new_dir)->i_projid,
 			F2FS_I(old_dentry->d_inode)->i_projid)) ||
@@ -1124,6 +1091,9 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	}
 	f2fs_mark_inode_dirty_sync(new_dir, false);
 
+	add_ino_entry(sbi, old_dir->i_ino, TRANS_DIR_INO);
+	add_ino_entry(sbi, new_dir->i_ino, TRANS_DIR_INO);
+
 	f2fs_unlock_op(sbi);
 
 	if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir))
@@ -1153,9 +1123,16 @@ static int f2fs_rename2(struct inode *old_dir, struct dentry *old_dentry,
 			struct inode *new_dir, struct dentry *new_dentry,
 			unsigned int flags)
 {
+	int err;
+
 	if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT))
 		return -EINVAL;
 
+	err = fscrypt_prepare_rename(old_dir, old_dentry, new_dir, new_dentry,
+				     flags);
+	if (err)
+		return err;
+
 	if (flags & RENAME_EXCHANGE) {
 		return f2fs_cross_rename(old_dir, old_dentry,
 					 new_dir, new_dentry);

fs/f2fs/node.c

@@ -143,11 +143,9 @@ static struct nat_entry *__alloc_nat_entry(nid_t nid, bool no_fail)
 	struct nat_entry *new;
 
 	if (no_fail)
-		new = f2fs_kmem_cache_alloc(nat_entry_slab,
-						GFP_NOFS | __GFP_ZERO);
+		new = f2fs_kmem_cache_alloc(nat_entry_slab, GFP_F2FS_ZERO);
 	else
-		new = kmem_cache_alloc(nat_entry_slab,
-						GFP_NOFS | __GFP_ZERO);
+		new = kmem_cache_alloc(nat_entry_slab, GFP_F2FS_ZERO);
 	if (new) {
 		nat_set_nid(new, nid);
 		nat_reset_flag(new);
@@ -702,7 +700,6 @@ static void truncate_node(struct dnode_of_data *dn)
 	struct node_info ni;
 
 	get_node_info(sbi, dn->nid, &ni);
-	f2fs_bug_on(sbi, ni.blk_addr == NULL_ADDR);
 
 	/* Deallocate node address */
 	invalidate_blocks(sbi, ni.blk_addr);
@@ -1336,14 +1333,19 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted,
 		.encrypted_page = NULL,
 		.submitted = false,
 		.io_type = io_type,
+		.io_wbc = wbc,
 	};
 
 	trace_f2fs_writepage(page, NODE);
 
+	if (unlikely(f2fs_cp_error(sbi))) {
+		dec_page_count(sbi, F2FS_DIRTY_NODES);
+		unlock_page(page);
+		return 0;
+	}
+
 	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
 		goto redirty_out;
-	if (unlikely(f2fs_cp_error(sbi)))
-		goto redirty_out;
 
 	/* get old block addr of this node page */
 	nid = nid_of_node(page);
@@ -1580,12 +1582,6 @@ next_step:
 			struct page *page = pvec.pages[i];
 			bool submitted = false;
 
-			if (unlikely(f2fs_cp_error(sbi))) {
-				pagevec_release(&pvec);
-				ret = -EIO;
-				goto out;
-			}
-
 			/*
 			 * flushing sequence with step:
 			 * 0. indirect nodes
@@ -1655,9 +1651,12 @@ continue_unlock:
 		step++;
 		goto next_step;
 	}
-out:
+
 	if (nwritten)
 		f2fs_submit_merged_write(sbi, NODE);
+
+	if (unlikely(f2fs_cp_error(sbi)))
+		return -EIO;
 	return ret;
 }
 
@@ -1812,8 +1811,33 @@ static void __move_free_nid(struct f2fs_sb_info *sbi, struct free_nid *i,
 	}
 }
 
+static void update_free_nid_bitmap(struct f2fs_sb_info *sbi, nid_t nid,
+							bool set, bool build)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	unsigned int nat_ofs = NAT_BLOCK_OFFSET(nid);
+	unsigned int nid_ofs = nid - START_NID(nid);
+
+	if (!test_bit_le(nat_ofs, nm_i->nat_block_bitmap))
+		return;
+
+	if (set) {
+		if (test_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]))
+			return;
+		__set_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]);
+		nm_i->free_nid_count[nat_ofs]++;
+	} else {
+		if (!test_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]))
+			return;
+		__clear_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]);
+		if (!build)
+			nm_i->free_nid_count[nat_ofs]--;
+	}
+}
+
 /* return if the nid is recognized as free */
-static bool add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
+static bool add_free_nid(struct f2fs_sb_info *sbi,
+				nid_t nid, bool build, bool update)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct free_nid *i, *e;
@@ -1829,8 +1853,7 @@ static bool add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
 	i->nid = nid;
 	i->state = FREE_NID;
 
-	if (radix_tree_preload(GFP_NOFS))
-		goto err;
+	radix_tree_preload(GFP_NOFS | __GFP_NOFAIL);
 
 	spin_lock(&nm_i->nid_list_lock);
 
@@ -1871,9 +1894,14 @@ static bool add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
 	ret = true;
 	err = __insert_free_nid(sbi, i, FREE_NID);
 err_out:
+	if (update) {
+		update_free_nid_bitmap(sbi, nid, ret, build);
+		if (!build)
+			nm_i->available_nids++;
+	}
 	spin_unlock(&nm_i->nid_list_lock);
 	radix_tree_preload_end();
-err:
+
 	if (err)
 		kmem_cache_free(free_nid_slab, i);
 	return ret;
@@ -1897,30 +1925,6 @@ static void remove_free_nid(struct f2fs_sb_info *sbi, nid_t nid)
 		kmem_cache_free(free_nid_slab, i);
 }
 
-static void update_free_nid_bitmap(struct f2fs_sb_info *sbi, nid_t nid,
-							bool set, bool build)
-{
-	struct f2fs_nm_info *nm_i = NM_I(sbi);
-	unsigned int nat_ofs = NAT_BLOCK_OFFSET(nid);
-	unsigned int nid_ofs = nid - START_NID(nid);
-
-	if (!test_bit_le(nat_ofs, nm_i->nat_block_bitmap))
-		return;
-
-	if (set) {
-		if (test_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]))
-			return;
-		__set_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]);
-		nm_i->free_nid_count[nat_ofs]++;
-	} else {
-		if (!test_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]))
-			return;
-		__clear_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]);
-		if (!build)
-			nm_i->free_nid_count[nat_ofs]--;
-	}
-}
-
 static void scan_nat_page(struct f2fs_sb_info *sbi,
 			struct page *nat_page, nid_t start_nid)
 {
@@ -1930,26 +1934,23 @@ static void scan_nat_page(struct f2fs_sb_info *sbi,
 	unsigned int nat_ofs = NAT_BLOCK_OFFSET(start_nid);
 	int i;
 
-	if (test_bit_le(nat_ofs, nm_i->nat_block_bitmap))
-		return;
-
 	__set_bit_le(nat_ofs, nm_i->nat_block_bitmap);
 
 	i = start_nid % NAT_ENTRY_PER_BLOCK;
 
 	for (; i < NAT_ENTRY_PER_BLOCK; i++, start_nid++) {
-		bool freed = false;
-
 		if (unlikely(start_nid >= nm_i->max_nid))
 			break;
 
 		blk_addr = le32_to_cpu(nat_blk->entries[i].block_addr);
 		f2fs_bug_on(sbi, blk_addr == NEW_ADDR);
-		if (blk_addr == NULL_ADDR)
-			freed = add_free_nid(sbi, start_nid, true);
-		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);
+		if (blk_addr == NULL_ADDR) {
+			add_free_nid(sbi, start_nid, true, true);
+		} else {
+			spin_lock(&NM_I(sbi)->nid_list_lock);
+			update_free_nid_bitmap(sbi, start_nid, false, true);
+			spin_unlock(&NM_I(sbi)->nid_list_lock);
+		}
 	}
 }
 
@@ -1967,7 +1968,7 @@ static void scan_curseg_cache(struct f2fs_sb_info *sbi)
 		addr = le32_to_cpu(nat_in_journal(journal, i).block_addr);
 		nid = le32_to_cpu(nid_in_journal(journal, i));
 		if (addr == NULL_ADDR)
-			add_free_nid(sbi, nid, true);
+			add_free_nid(sbi, nid, true, false);
 		else
 			remove_free_nid(sbi, nid);
 	}
@@ -1994,7 +1995,7 @@ static void scan_free_nid_bits(struct f2fs_sb_info *sbi)
 				break;
 
 			nid = i * NAT_ENTRY_PER_BLOCK + idx;
-			add_free_nid(sbi, nid, true);
+			add_free_nid(sbi, nid, true, false);
 
 			if (nm_i->nid_cnt[FREE_NID] >= MAX_FREE_NIDS)
 				goto out;
@@ -2037,10 +2038,13 @@ static void __build_free_nids(struct f2fs_sb_info *sbi, bool sync, bool mount)
 	down_read(&nm_i->nat_tree_lock);
 
 	while (1) {
-		struct page *page = get_current_nat_page(sbi, nid);
+		if (!test_bit_le(NAT_BLOCK_OFFSET(nid),
+						nm_i->nat_block_bitmap)) {
+			struct page *page = get_current_nat_page(sbi, nid);
 
-		scan_nat_page(sbi, page, nid);
-		f2fs_put_page(page, 1);
+			scan_nat_page(sbi, page, nid);
+			f2fs_put_page(page, 1);
+		}
 
 		nid += (NAT_ENTRY_PER_BLOCK - (nid % NAT_ENTRY_PER_BLOCK));
 		if (unlikely(nid >= nm_i->max_nid))
@@ -2203,7 +2207,9 @@ void recover_inline_xattr(struct inode *inode, struct page *page)
 	f2fs_bug_on(F2FS_I_SB(inode), IS_ERR(ipage));
 
 	ri = F2FS_INODE(page);
-	if (!(ri->i_inline & F2FS_INLINE_XATTR)) {
+	if (ri->i_inline & F2FS_INLINE_XATTR) {
+		set_inode_flag(inode, FI_INLINE_XATTR);
+	} else {
 		clear_inode_flag(inode, FI_INLINE_XATTR);
 		goto update_inode;
 	}
@@ -2219,7 +2225,7 @@ update_inode:
 	f2fs_put_page(ipage, 1);
 }
 
-int recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
+int recover_xattr_data(struct inode *inode, struct page *page)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	nid_t prev_xnid = F2FS_I(inode)->i_xattr_nid;
@@ -2233,7 +2239,6 @@ int recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
 
 	/* 1: invalidate the previous xattr nid */
 	get_node_info(sbi, prev_xnid, &ni);
-	f2fs_bug_on(sbi, ni.blk_addr == NULL_ADDR);
 	invalidate_blocks(sbi, ni.blk_addr);
 	dec_valid_node_count(sbi, inode, false);
 	set_node_addr(sbi, &ni, NULL_ADDR, false);
@@ -2322,7 +2327,7 @@ retry:
 	return 0;
 }
 
-int restore_node_summary(struct f2fs_sb_info *sbi,
+void restore_node_summary(struct f2fs_sb_info *sbi,
 			unsigned int segno, struct f2fs_summary_block *sum)
 {
 	struct f2fs_node *rn;
@@ -2355,7 +2360,6 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
 		invalidate_mapping_pages(META_MAPPING(sbi), addr,
 							addr + nrpages);
 	}
-	return 0;
 }
 
 static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
@@ -2497,11 +2501,7 @@ static void __flush_nat_entry_set(struct f2fs_sb_info *sbi,
 		nat_reset_flag(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);
-			spin_unlock(&NM_I(sbi)->nid_list_lock);
+			add_free_nid(sbi, nid, false, true);
 		} else {
 			spin_lock(&NM_I(sbi)->nid_list_lock);
 			update_free_nid_bitmap(sbi, nid, false, false);
@@ -2582,8 +2582,8 @@ static int __get_nat_bitmaps(struct f2fs_sb_info *sbi)
 
 	nm_i->nat_bits_blocks = F2FS_BYTES_TO_BLK((nat_bits_bytes << 1) + 8 +
 						F2FS_BLKSIZE - 1);
-	nm_i->nat_bits = kzalloc(nm_i->nat_bits_blocks << F2FS_BLKSIZE_BITS,
-						GFP_KERNEL);
+	nm_i->nat_bits = f2fs_kzalloc(sbi,
+			nm_i->nat_bits_blocks << F2FS_BLKSIZE_BITS, GFP_KERNEL);
 	if (!nm_i->nat_bits)
 		return -ENOMEM;
 
@@ -2661,7 +2661,7 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
 
 	/* not used nids: 0, node, meta, (and root counted as valid node) */
 	nm_i->available_nids = nm_i->max_nid - sbi->total_valid_node_count -
-							F2FS_RESERVED_NODE_NUM;
+				sbi->nquota_files - F2FS_RESERVED_NODE_NUM;
 	nm_i->nid_cnt[FREE_NID] = 0;
 	nm_i->nid_cnt[PREALLOC_NID] = 0;
 	nm_i->nat_cnt = 0;
@@ -2708,17 +2708,17 @@ static int init_free_nid_cache(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 
-	nm_i->free_nid_bitmap = kvzalloc(nm_i->nat_blocks *
+	nm_i->free_nid_bitmap = f2fs_kvzalloc(sbi, nm_i->nat_blocks *
 					NAT_ENTRY_BITMAP_SIZE, GFP_KERNEL);
 	if (!nm_i->free_nid_bitmap)
 		return -ENOMEM;
 
-	nm_i->nat_block_bitmap = kvzalloc(nm_i->nat_blocks / 8,
+	nm_i->nat_block_bitmap = f2fs_kvzalloc(sbi, nm_i->nat_blocks / 8,
 								GFP_KERNEL);
 	if (!nm_i->nat_block_bitmap)
 		return -ENOMEM;
 
-	nm_i->free_nid_count = kvzalloc(nm_i->nat_blocks *
+	nm_i->free_nid_count = f2fs_kvzalloc(sbi, nm_i->nat_blocks *
 					sizeof(unsigned short), GFP_KERNEL);
 	if (!nm_i->free_nid_count)
 		return -ENOMEM;
@@ -2729,7 +2729,8 @@ int build_node_manager(struct f2fs_sb_info *sbi)
 {
 	int err;
 
-	sbi->nm_info = kzalloc(sizeof(struct f2fs_nm_info), GFP_KERNEL);
+	sbi->nm_info = f2fs_kzalloc(sbi, sizeof(struct f2fs_nm_info),
+							GFP_KERNEL);
 	if (!sbi->nm_info)
 		return -ENOMEM;
 

fs/f2fs/node.h

@@ -305,6 +305,10 @@ static inline bool is_recoverable_dnode(struct page *page)
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(F2FS_P_SB(page));
 	__u64 cp_ver = cur_cp_version(ckpt);
 
+	/* Don't care crc part, if fsck.f2fs sets it. */
+	if (__is_set_ckpt_flags(ckpt, CP_NOCRC_RECOVERY_FLAG))
+		return (cp_ver << 32) == (cpver_of_node(page) << 32);
+
 	if (__is_set_ckpt_flags(ckpt, CP_CRC_RECOVERY_FLAG))
 		cp_ver |= (cur_cp_crc(ckpt) << 32);
 

fs/f2fs/recovery.c

@@ -195,6 +195,20 @@ out:
 	return err;
 }
 
+static void recover_inline_flags(struct inode *inode, struct f2fs_inode *ri)
+{
+	if (ri->i_inline & F2FS_PIN_FILE)
+		set_inode_flag(inode, FI_PIN_FILE);
+	else
+		clear_inode_flag(inode, FI_PIN_FILE);
+	if (ri->i_inline & F2FS_DATA_EXIST)
+		set_inode_flag(inode, FI_DATA_EXIST);
+	else
+		clear_inode_flag(inode, FI_DATA_EXIST);
+	if (!(ri->i_inline & F2FS_INLINE_DOTS))
+		clear_inode_flag(inode, FI_INLINE_DOTS);
+}
+
 static void recover_inode(struct inode *inode, struct page *page)
 {
 	struct f2fs_inode *raw = F2FS_INODE(page);
@@ -211,13 +225,16 @@ static void recover_inode(struct inode *inode, struct page *page)
 
 	F2FS_I(inode)->i_advise = raw->i_advise;
 
+	recover_inline_flags(inode, raw);
+
 	if (file_enc_name(inode))
 		name = "<encrypted>";
 	else
 		name = F2FS_INODE(page)->i_name;
 
-	f2fs_msg(inode->i_sb, KERN_NOTICE, "recover_inode: ino = %x, name = %s",
-			ino_of_node(page), name);
+	f2fs_msg(inode->i_sb, KERN_NOTICE,
+		"recover_inode: ino = %x, name = %s, inline = %x",
+			ino_of_node(page), name, raw->i_inline);
 }
 
 static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
@@ -404,7 +421,7 @@ truncate_out:
 }
 
 static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
-					struct page *page, block_t blkaddr)
+					struct page *page)
 {
 	struct dnode_of_data dn;
 	struct node_info ni;
@@ -415,7 +432,7 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 	if (IS_INODE(page)) {
 		recover_inline_xattr(inode, page);
 	} else if (f2fs_has_xattr_block(ofs_of_node(page))) {
-		err = recover_xattr_data(inode, page, blkaddr);
+		err = recover_xattr_data(inode, page);
 		if (!err)
 			recovered++;
 		goto out;
@@ -568,7 +585,7 @@ static int recover_data(struct f2fs_sb_info *sbi, struct list_head *inode_list,
 				break;
 			}
 		}
-		err = do_recover_data(sbi, entry->inode, page, blkaddr);
+		err = do_recover_data(sbi, entry->inode, page);
 		if (err) {
 			f2fs_put_page(page, 1);
 			break;

fs/f2fs/segment.c

@@ -248,7 +248,11 @@ retry:
 				goto next;
 			}
 			get_node_info(sbi, dn.nid, &ni);
-			f2fs_replace_block(sbi, &dn, dn.data_blkaddr,
+			if (cur->old_addr == NEW_ADDR) {
+				invalidate_blocks(sbi, dn.data_blkaddr);
+				f2fs_update_data_blkaddr(&dn, NEW_ADDR);
+			} else
+				f2fs_replace_block(sbi, &dn, dn.data_blkaddr,
 					cur->old_addr, ni.version, true, true);
 			f2fs_put_dnode(&dn);
 		}
@@ -657,7 +661,7 @@ int create_flush_cmd_control(struct f2fs_sb_info *sbi)
 		goto init_thread;
 	}
 
-	fcc = kzalloc(sizeof(struct flush_cmd_control), GFP_KERNEL);
+	fcc = f2fs_kzalloc(sbi, sizeof(struct flush_cmd_control), GFP_KERNEL);
 	if (!fcc)
 		return -ENOMEM;
 	atomic_set(&fcc->issued_flush, 0);
@@ -884,7 +888,7 @@ static void f2fs_submit_discard_endio(struct bio *bio)
 	bio_put(bio);
 }
 
-void __check_sit_bitmap(struct f2fs_sb_info *sbi,
+static void __check_sit_bitmap(struct f2fs_sb_info *sbi,
 				block_t start, block_t end)
 {
 #ifdef CONFIG_F2FS_CHECK_FS
@@ -1204,6 +1208,8 @@ static int __issue_discard_cmd(struct f2fs_sb_info *sbi,
 		pend_list = &dcc->pend_list[i];
 
 		mutex_lock(&dcc->cmd_lock);
+		if (list_empty(pend_list))
+			goto next;
 		f2fs_bug_on(sbi, !__check_rb_tree_consistence(sbi, &dcc->root));
 		blk_start_plug(&plug);
 		list_for_each_entry_safe(dc, tmp, pend_list, list) {
@@ -1222,6 +1228,7 @@ skip:
 				break;
 		}
 		blk_finish_plug(&plug);
+next:
 		mutex_unlock(&dcc->cmd_lock);
 
 		if (iter >= dpolicy->max_requests)
@@ -1256,6 +1263,11 @@ static bool __drop_discard_cmd(struct f2fs_sb_info *sbi)
 	return dropped;
 }
 
+void drop_discard_cmd(struct f2fs_sb_info *sbi)
+{
+	__drop_discard_cmd(sbi);
+}
+
 static unsigned int __wait_one_discard_bio(struct f2fs_sb_info *sbi,
 							struct discard_cmd *dc)
 {
@@ -1324,7 +1336,7 @@ static void __wait_all_discard_cmd(struct f2fs_sb_info *sbi,
 }
 
 /* This should be covered by global mutex, &sit_i->sentry_lock */
-void f2fs_wait_discard_bio(struct f2fs_sb_info *sbi, block_t blkaddr)
+static void f2fs_wait_discard_bio(struct f2fs_sb_info *sbi, block_t blkaddr)
 {
 	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
 	struct discard_cmd *dc;
@@ -1394,6 +1406,8 @@ static int issue_discard_thread(void *data)
 				msecs_to_jiffies(wait_ms));
 		if (try_to_freeze())
 			continue;
+		if (f2fs_readonly(sbi->sb))
+			continue;
 		if (kthread_should_stop())
 			return 0;
 
@@ -1703,25 +1717,20 @@ void init_discard_policy(struct discard_policy *dpolicy,
 	dpolicy->sync = true;
 	dpolicy->granularity = granularity;
 
+	dpolicy->max_requests = DEF_MAX_DISCARD_REQUEST;
+	dpolicy->io_aware_gran = MAX_PLIST_NUM;
+
 	if (discard_type == DPOLICY_BG) {
 		dpolicy->min_interval = DEF_MIN_DISCARD_ISSUE_TIME;
 		dpolicy->max_interval = DEF_MAX_DISCARD_ISSUE_TIME;
-		dpolicy->max_requests = DEF_MAX_DISCARD_REQUEST;
-		dpolicy->io_aware_gran = MAX_PLIST_NUM;
 		dpolicy->io_aware = true;
 	} else if (discard_type == DPOLICY_FORCE) {
 		dpolicy->min_interval = DEF_MIN_DISCARD_ISSUE_TIME;
 		dpolicy->max_interval = DEF_MAX_DISCARD_ISSUE_TIME;
-		dpolicy->max_requests = DEF_MAX_DISCARD_REQUEST;
-		dpolicy->io_aware_gran = MAX_PLIST_NUM;
 		dpolicy->io_aware = true;
 	} else if (discard_type == DPOLICY_FSTRIM) {
-		dpolicy->max_requests = DEF_MAX_DISCARD_REQUEST;
-		dpolicy->io_aware_gran = MAX_PLIST_NUM;
 		dpolicy->io_aware = false;
 	} else if (discard_type == DPOLICY_UMOUNT) {
-		dpolicy->max_requests = DEF_MAX_DISCARD_REQUEST;
-		dpolicy->io_aware_gran = MAX_PLIST_NUM;
 		dpolicy->io_aware = false;
 	}
 }
@@ -1737,7 +1746,7 @@ static int create_discard_cmd_control(struct f2fs_sb_info *sbi)
 		goto init_thread;
 	}
 
-	dcc = kzalloc(sizeof(struct discard_cmd_control), GFP_KERNEL);
+	dcc = f2fs_kzalloc(sbi, sizeof(struct discard_cmd_control), GFP_KERNEL);
 	if (!dcc)
 		return -ENOMEM;
 
@@ -2739,6 +2748,7 @@ void __f2fs_replace_block(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
 		}
 	}
 
+	f2fs_bug_on(sbi, !IS_DATASEG(type));
 	curseg = CURSEG_I(sbi, type);
 
 	mutex_lock(&curseg->curseg_mutex);
@@ -2823,7 +2833,7 @@ void f2fs_wait_on_block_writeback(struct f2fs_sb_info *sbi, block_t blkaddr)
 	}
 }
 
-static int read_compacted_summaries(struct f2fs_sb_info *sbi)
+static void read_compacted_summaries(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
 	struct curseg_info *seg_i;
@@ -2880,7 +2890,6 @@ static int read_compacted_summaries(struct f2fs_sb_info *sbi)
 		}
 	}
 	f2fs_put_page(page, 1);
-	return 0;
 }
 
 static int read_normal_summaries(struct f2fs_sb_info *sbi, int type)
@@ -2926,13 +2935,7 @@ static int read_normal_summaries(struct f2fs_sb_info *sbi, int type)
 				ns->ofs_in_node = 0;
 			}
 		} else {
-			int err;
-
-			err = restore_node_summary(sbi, segno, sum);
-			if (err) {
-				f2fs_put_page(new, 1);
-				return err;
-			}
+			restore_node_summary(sbi, segno, sum);
 		}
 	}
 
@@ -2971,8 +2974,7 @@ static int restore_curseg_summaries(struct f2fs_sb_info *sbi)
 							META_CP, true);
 
 		/* restore for compacted data summary */
-		if (read_compacted_summaries(sbi))
-			return -EINVAL;
+		read_compacted_summaries(sbi);
 		type = CURSEG_HOT_NODE;
 	}
 
@@ -3108,28 +3110,19 @@ static struct page *get_next_sit_page(struct f2fs_sb_info *sbi,
 					unsigned int start)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	struct page *src_page, *dst_page;
+	struct page *page;
 	pgoff_t src_off, dst_off;
-	void *src_addr, *dst_addr;
 
 	src_off = current_sit_addr(sbi, start);
 	dst_off = next_sit_addr(sbi, src_off);
 
-	/* get current sit block page without lock */
-	src_page = get_meta_page(sbi, src_off);
-	dst_page = grab_meta_page(sbi, dst_off);
-	f2fs_bug_on(sbi, PageDirty(src_page));
-
-	src_addr = page_address(src_page);
-	dst_addr = page_address(dst_page);
-	memcpy(dst_addr, src_addr, PAGE_SIZE);
-
-	set_page_dirty(dst_page);
-	f2fs_put_page(src_page, 1);
+	page = grab_meta_page(sbi, dst_off);
+	seg_info_to_sit_page(sbi, page, start);
 
+	set_page_dirty(page);
 	set_to_next_sit(sit_i, start);
 
-	return dst_page;
+	return page;
 }
 
 static struct sit_entry_set *grab_sit_entry_set(void)
@@ -3338,52 +3331,54 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 	unsigned int bitmap_size;
 
 	/* allocate memory for SIT information */
-	sit_i = kzalloc(sizeof(struct sit_info), GFP_KERNEL);
+	sit_i = f2fs_kzalloc(sbi, sizeof(struct sit_info), GFP_KERNEL);
 	if (!sit_i)
 		return -ENOMEM;
 
 	SM_I(sbi)->sit_info = sit_i;
 
-	sit_i->sentries = kvzalloc(MAIN_SEGS(sbi) *
+	sit_i->sentries = f2fs_kvzalloc(sbi, MAIN_SEGS(sbi) *
 					sizeof(struct seg_entry), GFP_KERNEL);
 	if (!sit_i->sentries)
 		return -ENOMEM;
 
 	bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
-	sit_i->dirty_sentries_bitmap = kvzalloc(bitmap_size, GFP_KERNEL);
+	sit_i->dirty_sentries_bitmap = f2fs_kvzalloc(sbi, bitmap_size,
+								GFP_KERNEL);
 	if (!sit_i->dirty_sentries_bitmap)
 		return -ENOMEM;
 
 	for (start = 0; start < MAIN_SEGS(sbi); start++) {
 		sit_i->sentries[start].cur_valid_map
-			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+			= f2fs_kzalloc(sbi, SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 		sit_i->sentries[start].ckpt_valid_map
-			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+			= f2fs_kzalloc(sbi, SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 		if (!sit_i->sentries[start].cur_valid_map ||
 				!sit_i->sentries[start].ckpt_valid_map)
 			return -ENOMEM;
 
 #ifdef CONFIG_F2FS_CHECK_FS
 		sit_i->sentries[start].cur_valid_map_mir
-			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+			= f2fs_kzalloc(sbi, SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 		if (!sit_i->sentries[start].cur_valid_map_mir)
 			return -ENOMEM;
 #endif
 
 		if (f2fs_discard_en(sbi)) {
 			sit_i->sentries[start].discard_map
-				= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+				= f2fs_kzalloc(sbi, SIT_VBLOCK_MAP_SIZE,
+								GFP_KERNEL);
 			if (!sit_i->sentries[start].discard_map)
 				return -ENOMEM;
 		}
 	}
 
-	sit_i->tmp_map = kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+	sit_i->tmp_map = f2fs_kzalloc(sbi, SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 	if (!sit_i->tmp_map)
 		return -ENOMEM;
 
 	if (sbi->segs_per_sec > 1) {
-		sit_i->sec_entries = kvzalloc(MAIN_SECS(sbi) *
+		sit_i->sec_entries = f2fs_kvzalloc(sbi, MAIN_SECS(sbi) *
 					sizeof(struct sec_entry), GFP_KERNEL);
 		if (!sit_i->sec_entries)
 			return -ENOMEM;
@@ -3427,19 +3422,19 @@ static int build_free_segmap(struct f2fs_sb_info *sbi)
 	unsigned int bitmap_size, sec_bitmap_size;
 
 	/* allocate memory for free segmap information */
-	free_i = kzalloc(sizeof(struct free_segmap_info), GFP_KERNEL);
+	free_i = f2fs_kzalloc(sbi, sizeof(struct free_segmap_info), GFP_KERNEL);
 	if (!free_i)
 		return -ENOMEM;
 
 	SM_I(sbi)->free_info = free_i;
 
 	bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
-	free_i->free_segmap = kvmalloc(bitmap_size, GFP_KERNEL);
+	free_i->free_segmap = f2fs_kvmalloc(sbi, bitmap_size, GFP_KERNEL);
 	if (!free_i->free_segmap)
 		return -ENOMEM;
 
 	sec_bitmap_size = f2fs_bitmap_size(MAIN_SECS(sbi));
-	free_i->free_secmap = kvmalloc(sec_bitmap_size, GFP_KERNEL);
+	free_i->free_secmap = f2fs_kvmalloc(sbi, sec_bitmap_size, GFP_KERNEL);
 	if (!free_i->free_secmap)
 		return -ENOMEM;
 
@@ -3460,7 +3455,7 @@ static int build_curseg(struct f2fs_sb_info *sbi)
 	struct curseg_info *array;
 	int i;
 
-	array = kcalloc(NR_CURSEG_TYPE, sizeof(*array), GFP_KERNEL);
+	array = f2fs_kzalloc(sbi, sizeof(*array) * NR_CURSEG_TYPE, GFP_KERNEL);
 	if (!array)
 		return -ENOMEM;
 
@@ -3468,12 +3463,12 @@ static int build_curseg(struct f2fs_sb_info *sbi)
 
 	for (i = 0; i < NR_CURSEG_TYPE; i++) {
 		mutex_init(&array[i].curseg_mutex);
-		array[i].sum_blk = kzalloc(PAGE_SIZE, GFP_KERNEL);
+		array[i].sum_blk = f2fs_kzalloc(sbi, PAGE_SIZE, GFP_KERNEL);
 		if (!array[i].sum_blk)
 			return -ENOMEM;
 		init_rwsem(&array[i].journal_rwsem);
-		array[i].journal = kzalloc(sizeof(struct f2fs_journal),
-							GFP_KERNEL);
+		array[i].journal = f2fs_kzalloc(sbi,
+				sizeof(struct f2fs_journal), GFP_KERNEL);
 		if (!array[i].journal)
 			return -ENOMEM;
 		array[i].segno = NULL_SEGNO;
@@ -3482,7 +3477,7 @@ static int build_curseg(struct f2fs_sb_info *sbi)
 	return restore_curseg_summaries(sbi);
 }
 
-static void build_sit_entries(struct f2fs_sb_info *sbi)
+static int build_sit_entries(struct f2fs_sb_info *sbi)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
@@ -3492,6 +3487,7 @@ static void build_sit_entries(struct f2fs_sb_info *sbi)
 	int sit_blk_cnt = SIT_BLK_CNT(sbi);
 	unsigned int i, start, end;
 	unsigned int readed, start_blk = 0;
+	int err = 0;
 
 	do {
 		readed = ra_meta_pages(sbi, start_blk, BIO_MAX_PAGES,
@@ -3510,7 +3506,9 @@ static void build_sit_entries(struct f2fs_sb_info *sbi)
 			sit = sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, start)];
 			f2fs_put_page(page, 1);
 
-			check_block_count(sbi, start, &sit);
+			err = check_block_count(sbi, start, &sit);
+			if (err)
+				return err;
 			seg_info_from_raw_sit(se, &sit);
 
 			/* build discard map only one time */
@@ -3545,7 +3543,9 @@ static void build_sit_entries(struct f2fs_sb_info *sbi)
 
 		old_valid_blocks = se->valid_blocks;
 
-		check_block_count(sbi, start, &sit);
+		err = check_block_count(sbi, start, &sit);
+		if (err)
+			break;
 		seg_info_from_raw_sit(se, &sit);
 
 		if (f2fs_discard_en(sbi)) {
@@ -3565,6 +3565,7 @@ static void build_sit_entries(struct f2fs_sb_info *sbi)
 				se->valid_blocks - old_valid_blocks;
 	}
 	up_read(&curseg->journal_rwsem);
+	return err;
 }
 
 static void init_free_segmap(struct f2fs_sb_info *sbi)
@@ -3619,7 +3620,7 @@ static int init_victim_secmap(struct f2fs_sb_info *sbi)
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
 	unsigned int bitmap_size = f2fs_bitmap_size(MAIN_SECS(sbi));
 
-	dirty_i->victim_secmap = kvzalloc(bitmap_size, GFP_KERNEL);
+	dirty_i->victim_secmap = f2fs_kvzalloc(sbi, bitmap_size, GFP_KERNEL);
 	if (!dirty_i->victim_secmap)
 		return -ENOMEM;
 	return 0;
@@ -3631,7 +3632,8 @@ static int build_dirty_segmap(struct f2fs_sb_info *sbi)
 	unsigned int bitmap_size, i;
 
 	/* allocate memory for dirty segments list information */
-	dirty_i = kzalloc(sizeof(struct dirty_seglist_info), GFP_KERNEL);
+	dirty_i = f2fs_kzalloc(sbi, sizeof(struct dirty_seglist_info),
+								GFP_KERNEL);
 	if (!dirty_i)
 		return -ENOMEM;
 
@@ -3641,7 +3643,8 @@ static int build_dirty_segmap(struct f2fs_sb_info *sbi)
 	bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
 
 	for (i = 0; i < NR_DIRTY_TYPE; i++) {
-		dirty_i->dirty_segmap[i] = kvzalloc(bitmap_size, GFP_KERNEL);
+		dirty_i->dirty_segmap[i] = f2fs_kvzalloc(sbi, bitmap_size,
+								GFP_KERNEL);
 		if (!dirty_i->dirty_segmap[i])
 			return -ENOMEM;
 	}
@@ -3685,7 +3688,7 @@ int build_segment_manager(struct f2fs_sb_info *sbi)
 	struct f2fs_sm_info *sm_info;
 	int err;
 
-	sm_info = kzalloc(sizeof(struct f2fs_sm_info), GFP_KERNEL);
+	sm_info = f2fs_kzalloc(sbi, sizeof(struct f2fs_sm_info), GFP_KERNEL);
 	if (!sm_info)
 		return -ENOMEM;
 
@@ -3737,7 +3740,9 @@ int build_segment_manager(struct f2fs_sb_info *sbi)
 		return err;
 
 	/* reinit free segmap based on SIT */
-	build_sit_entries(sbi);
+	err = build_sit_entries(sbi);
+	if (err)
+		return err;
 
 	init_free_segmap(sbi);
 	err = build_dirty_segmap(sbi);

fs/f2fs/segment.h

@@ -348,16 +348,41 @@ static inline void seg_info_from_raw_sit(struct seg_entry *se,
 	se->mtime = le64_to_cpu(rs->mtime);
 }
 
-static inline void seg_info_to_raw_sit(struct seg_entry *se,
+static inline void __seg_info_to_raw_sit(struct seg_entry *se,
 					struct f2fs_sit_entry *rs)
 {
 	unsigned short raw_vblocks = (se->type << SIT_VBLOCKS_SHIFT) |
 					se->valid_blocks;
 	rs->vblocks = cpu_to_le16(raw_vblocks);
 	memcpy(rs->valid_map, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
+	rs->mtime = cpu_to_le64(se->mtime);
+}
+
+static inline void seg_info_to_sit_page(struct f2fs_sb_info *sbi,
+				struct page *page, unsigned int start)
+{
+	struct f2fs_sit_block *raw_sit;
+	struct seg_entry *se;
+	struct f2fs_sit_entry *rs;
+	unsigned int end = min(start + SIT_ENTRY_PER_BLOCK,
+					(unsigned long)MAIN_SEGS(sbi));
+	int i;
+
+	raw_sit = (struct f2fs_sit_block *)page_address(page);
+	for (i = 0; i < end - start; i++) {
+		rs = &raw_sit->entries[i];
+		se = get_seg_entry(sbi, start + i);
+		__seg_info_to_raw_sit(se, rs);
+	}
+}
+
+static inline void seg_info_to_raw_sit(struct seg_entry *se,
+					struct f2fs_sit_entry *rs)
+{
+	__seg_info_to_raw_sit(se, rs);
+
 	memcpy(se->ckpt_valid_map, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
 	se->ckpt_valid_blocks = se->valid_blocks;
-	rs->mtime = cpu_to_le64(se->mtime);
 }
 
 static inline unsigned int find_next_inuse(struct free_segmap_info *free_i,
@@ -580,47 +605,6 @@ enum {
 	F2FS_IPU_ASYNC,
 };
 
-static inline bool need_inplace_update_policy(struct inode *inode,
-				struct f2fs_io_info *fio)
-{
-	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	unsigned int policy = SM_I(sbi)->ipu_policy;
-
-	if (test_opt(sbi, LFS))
-		return false;
-
-	/* if this is cold file, we should overwrite to avoid fragmentation */
-	if (file_is_cold(inode))
-		return true;
-
-	if (policy & (0x1 << F2FS_IPU_FORCE))
-		return true;
-	if (policy & (0x1 << F2FS_IPU_SSR) && need_SSR(sbi))
-		return true;
-	if (policy & (0x1 << F2FS_IPU_UTIL) &&
-			utilization(sbi) > SM_I(sbi)->min_ipu_util)
-		return true;
-	if (policy & (0x1 << F2FS_IPU_SSR_UTIL) && need_SSR(sbi) &&
-			utilization(sbi) > SM_I(sbi)->min_ipu_util)
-		return true;
-
-	/*
-	 * IPU for rewrite async pages
-	 */
-	if (policy & (0x1 << F2FS_IPU_ASYNC) &&
-			fio && fio->op == REQ_OP_WRITE &&
-			!(fio->op_flags & REQ_SYNC) &&
-			!f2fs_encrypted_inode(inode))
-		return true;
-
-	/* this is only set during fdatasync */
-	if (policy & (0x1 << F2FS_IPU_FSYNC) &&
-			is_inode_flag_set(inode, FI_NEED_IPU))
-		return true;
-
-	return false;
-}
-
 static inline unsigned int curseg_segno(struct f2fs_sb_info *sbi,
 		int type)
 {
@@ -655,7 +639,7 @@ static inline void verify_block_addr(struct f2fs_sb_info *sbi, block_t blk_addr)
 /*
  * Summary block is always treated as an invalid block
  */
-static inline void check_block_count(struct f2fs_sb_info *sbi,
+static inline int check_block_count(struct f2fs_sb_info *sbi,
 		int segno, struct f2fs_sit_entry *raw_sit)
 {
 #ifdef CONFIG_F2FS_CHECK_FS
@@ -677,11 +661,25 @@ static inline void check_block_count(struct f2fs_sb_info *sbi,
 		cur_pos = next_pos;
 		is_valid = !is_valid;
 	} while (cur_pos < sbi->blocks_per_seg);
-	BUG_ON(GET_SIT_VBLOCKS(raw_sit) != valid_blocks);
+
+	if (unlikely(GET_SIT_VBLOCKS(raw_sit) != valid_blocks)) {
+		f2fs_msg(sbi->sb, KERN_ERR,
+				"Mismatch valid blocks %d vs. %d",
+					GET_SIT_VBLOCKS(raw_sit), valid_blocks);
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
+		return -EINVAL;
+	}
 #endif
 	/* check segment usage, and check boundary of a given segment number */
-	f2fs_bug_on(sbi, GET_SIT_VBLOCKS(raw_sit) > sbi->blocks_per_seg
-					|| segno > TOTAL_SEGS(sbi) - 1);
+	if (unlikely(GET_SIT_VBLOCKS(raw_sit) > sbi->blocks_per_seg
+					|| segno > TOTAL_SEGS(sbi) - 1)) {
+		f2fs_msg(sbi->sb, KERN_ERR,
+				"Wrong valid blocks %d or segno %u",
+					GET_SIT_VBLOCKS(raw_sit), segno);
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
+		return -EINVAL;
+	}
+	return 0;
 }
 
 static inline pgoff_t current_sit_addr(struct f2fs_sb_info *sbi,

fs/f2fs/super.c

@@ -43,6 +43,7 @@ static struct kmem_cache *f2fs_inode_cachep;
 
 char *fault_name[FAULT_MAX] = {
 	[FAULT_KMALLOC]		= "kmalloc",
+	[FAULT_KVMALLOC]	= "kvmalloc",
 	[FAULT_PAGE_ALLOC]	= "page alloc",
 	[FAULT_PAGE_GET]	= "page get",
 	[FAULT_ALLOC_BIO]	= "alloc bio",
@@ -106,6 +107,9 @@ enum {
 	Opt_noextent_cache,
 	Opt_noinline_data,
 	Opt_data_flush,
+	Opt_reserve_root,
+	Opt_resgid,
+	Opt_resuid,
 	Opt_mode,
 	Opt_io_size_bits,
 	Opt_fault_injection,
@@ -156,6 +160,9 @@ static match_table_t f2fs_tokens = {
 	{Opt_noextent_cache, "noextent_cache"},
 	{Opt_noinline_data, "noinline_data"},
 	{Opt_data_flush, "data_flush"},
+	{Opt_reserve_root, "reserve_root=%u"},
+	{Opt_resgid, "resgid=%u"},
+	{Opt_resuid, "resuid=%u"},
 	{Opt_mode, "mode=%s"},
 	{Opt_io_size_bits, "io_bits=%u"},
 	{Opt_fault_injection, "fault_injection=%u"},
@@ -190,6 +197,28 @@ void f2fs_msg(struct super_block *sb, const char *level, const char *fmt, ...)
 	va_end(args);
 }
 
+static inline void limit_reserve_root(struct f2fs_sb_info *sbi)
+{
+	block_t limit = (sbi->user_block_count << 1) / 1000;
+
+	/* limit is 0.2% */
+	if (test_opt(sbi, RESERVE_ROOT) && sbi->root_reserved_blocks > limit) {
+		sbi->root_reserved_blocks = limit;
+		f2fs_msg(sbi->sb, KERN_INFO,
+			"Reduce reserved blocks for root = %u",
+				sbi->root_reserved_blocks);
+	}
+	if (!test_opt(sbi, RESERVE_ROOT) &&
+		(!uid_eq(sbi->s_resuid,
+				make_kuid(&init_user_ns, F2FS_DEF_RESUID)) ||
+		!gid_eq(sbi->s_resgid,
+				make_kgid(&init_user_ns, F2FS_DEF_RESGID))))
+		f2fs_msg(sbi->sb, KERN_INFO,
+			"Ignore s_resuid=%u, s_resgid=%u w/o reserve_root",
+				from_kuid_munged(&init_user_ns, sbi->s_resuid),
+				from_kgid_munged(&init_user_ns, sbi->s_resgid));
+}
+
 static void init_once(void *foo)
 {
 	struct f2fs_inode_info *fi = (struct f2fs_inode_info *) foo;
@@ -320,6 +349,8 @@ static int parse_options(struct super_block *sb, char *options)
 	substring_t args[MAX_OPT_ARGS];
 	char *p, *name;
 	int arg = 0;
+	kuid_t uid;
+	kgid_t gid;
 #ifdef CONFIG_QUOTA
 	int ret;
 #endif
@@ -487,6 +518,40 @@ static int parse_options(struct super_block *sb, char *options)
 		case Opt_data_flush:
 			set_opt(sbi, DATA_FLUSH);
 			break;
+		case Opt_reserve_root:
+			if (args->from && match_int(args, &arg))
+				return -EINVAL;
+			if (test_opt(sbi, RESERVE_ROOT)) {
+				f2fs_msg(sb, KERN_INFO,
+					"Preserve previous reserve_root=%u",
+					sbi->root_reserved_blocks);
+			} else {
+				sbi->root_reserved_blocks = arg;
+				set_opt(sbi, RESERVE_ROOT);
+			}
+			break;
+		case Opt_resuid:
+			if (args->from && match_int(args, &arg))
+				return -EINVAL;
+			uid = make_kuid(current_user_ns(), arg);
+			if (!uid_valid(uid)) {
+				f2fs_msg(sb, KERN_ERR,
+					"Invalid uid value %d", arg);
+				return -EINVAL;
+			}
+			sbi->s_resuid = uid;
+			break;
+		case Opt_resgid:
+			if (args->from && match_int(args, &arg))
+				return -EINVAL;
+			gid = make_kgid(current_user_ns(), arg);
+			if (!gid_valid(gid)) {
+				f2fs_msg(sb, KERN_ERR,
+					"Invalid gid value %d", arg);
+				return -EINVAL;
+			}
+			sbi->s_resgid = gid;
+			break;
 		case Opt_mode:
 			name = match_strdup(&args[0]);
 
@@ -993,22 +1058,25 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	struct super_block *sb = dentry->d_sb;
 	struct f2fs_sb_info *sbi = F2FS_SB(sb);
 	u64 id = huge_encode_dev(sb->s_bdev->bd_dev);
-	block_t total_count, user_block_count, start_count, ovp_count;
+	block_t total_count, user_block_count, start_count;
 	u64 avail_node_count;
 
 	total_count = le64_to_cpu(sbi->raw_super->block_count);
 	user_block_count = sbi->user_block_count;
 	start_count = le32_to_cpu(sbi->raw_super->segment0_blkaddr);
-	ovp_count = SM_I(sbi)->ovp_segments << sbi->log_blocks_per_seg;
 	buf->f_type = F2FS_SUPER_MAGIC;
 	buf->f_bsize = sbi->blocksize;
 
 	buf->f_blocks = total_count - start_count;
-	buf->f_bfree = user_block_count - valid_user_blocks(sbi) + ovp_count;
-	buf->f_bavail = user_block_count - valid_user_blocks(sbi) -
+	buf->f_bfree = user_block_count - valid_user_blocks(sbi) -
 						sbi->current_reserved_blocks;
+	if (buf->f_bfree > sbi->root_reserved_blocks)
+		buf->f_bavail = buf->f_bfree - sbi->root_reserved_blocks;
+	else
+		buf->f_bavail = 0;
 
-	avail_node_count = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
+	avail_node_count = sbi->total_node_count - sbi->nquota_files -
+						F2FS_RESERVED_NODE_NUM;
 
 	if (avail_node_count > user_block_count) {
 		buf->f_files = user_block_count;
@@ -1134,6 +1202,11 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 	else if (test_opt(sbi, LFS))
 		seq_puts(seq, "lfs");
 	seq_printf(seq, ",active_logs=%u", sbi->active_logs);
+	if (test_opt(sbi, RESERVE_ROOT))
+		seq_printf(seq, ",reserve_root=%u,resuid=%u,resgid=%u",
+				sbi->root_reserved_blocks,
+				from_kuid_munged(&init_user_ns, sbi->s_resuid),
+				from_kgid_munged(&init_user_ns, sbi->s_resgid));
 	if (F2FS_IO_SIZE_BITS(sbi))
 		seq_printf(seq, ",io_size=%uKB", F2FS_IO_SIZE_KB(sbi));
 #ifdef CONFIG_F2FS_FAULT_INJECTION
@@ -1263,7 +1336,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 		err = dquot_suspend(sb, -1);
 		if (err < 0)
 			goto restore_opts;
-	} else {
+	} else if (f2fs_readonly(sb) && !(*flags & MS_RDONLY)) {
 		/* dquot_resume needs RW */
 		sb->s_flags &= ~SB_RDONLY;
 		if (sb_any_quota_suspended(sb)) {
@@ -1332,6 +1405,7 @@ skip:
 	sb->s_flags = (sb->s_flags & ~SB_POSIXACL) |
 		(test_opt(sbi, POSIX_ACL) ? SB_POSIXACL : 0);
 
+	limit_reserve_root(sbi);
 	return 0;
 restore_gc:
 	if (need_restart_gc) {
@@ -1656,7 +1730,7 @@ void f2fs_quota_off_umount(struct super_block *sb)
 		f2fs_quota_off(sb, type);
 }
 
-int f2fs_get_projid(struct inode *inode, kprojid_t *projid)
+static int f2fs_get_projid(struct inode *inode, kprojid_t *projid)
 {
 	*projid = F2FS_I(inode)->i_projid;
 	return 0;
@@ -2148,14 +2222,15 @@ static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
 	if (nr_sectors & (bdev_zone_sectors(bdev) - 1))
 		FDEV(devi).nr_blkz++;
 
-	FDEV(devi).blkz_type = kmalloc(FDEV(devi).nr_blkz, GFP_KERNEL);
+	FDEV(devi).blkz_type = f2fs_kmalloc(sbi, FDEV(devi).nr_blkz,
+								GFP_KERNEL);
 	if (!FDEV(devi).blkz_type)
 		return -ENOMEM;
 
 #define F2FS_REPORT_NR_ZONES   4096
 
-	zones = kcalloc(F2FS_REPORT_NR_ZONES, sizeof(struct blk_zone),
-			GFP_KERNEL);
+	zones = f2fs_kzalloc(sbi, sizeof(struct blk_zone) *
+				F2FS_REPORT_NR_ZONES, GFP_KERNEL);
 	if (!zones)
 		return -ENOMEM;
 
@@ -2295,8 +2370,8 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
 	 * Initialize multiple devices information, or single
 	 * zoned block device information.
 	 */
-	sbi->devs = kcalloc(max_devices, sizeof(struct f2fs_dev_info),
-				GFP_KERNEL);
+	sbi->devs = f2fs_kzalloc(sbi, sizeof(struct f2fs_dev_info) *
+						max_devices, GFP_KERNEL);
 	if (!sbi->devs)
 		return -ENOMEM;
 
@@ -2419,6 +2494,9 @@ try_onemore:
 	sb->s_fs_info = sbi;
 	sbi->raw_super = raw_super;
 
+	sbi->s_resuid = make_kuid(&init_user_ns, F2FS_DEF_RESUID);
+	sbi->s_resgid = make_kgid(&init_user_ns, F2FS_DEF_RESGID);
+
 	/* precompute checksum seed for metadata */
 	if (f2fs_sb_has_inode_chksum(sb))
 		sbi->s_chksum_seed = f2fs_chksum(sbi, ~0, raw_super->uuid,
@@ -2462,6 +2540,13 @@ try_onemore:
 	else
 		sb->s_qcop = &f2fs_quotactl_ops;
 	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP | QTYPE_MASK_PRJ;
+
+	if (f2fs_sb_has_quota_ino(sbi->sb)) {
+		for (i = 0; i < MAXQUOTAS; i++) {
+			if (f2fs_qf_ino(sbi->sb, i))
+				sbi->nquota_files++;
+		}
+	}
 #endif
 
 	sb->s_op = &f2fs_sops;
@@ -2475,6 +2560,7 @@ try_onemore:
 	sb->s_flags = (sb->s_flags & ~SB_POSIXACL) |
 		(test_opt(sbi, POSIX_ACL) ? SB_POSIXACL : 0);
 	memcpy(&sb->s_uuid, raw_super->uuid, sizeof(raw_super->uuid));
+	sb->s_iflags |= SB_I_CGROUPWB;
 
 	/* init f2fs-specific super block info */
 	sbi->valid_super_block = valid_super_block;
@@ -2495,8 +2581,9 @@ try_onemore:
 		int n = (i == META) ? 1: NR_TEMP_TYPE;
 		int j;
 
-		sbi->write_io[i] = kmalloc(n * sizeof(struct f2fs_bio_info),
-								GFP_KERNEL);
+		sbi->write_io[i] = f2fs_kmalloc(sbi,
+					n * sizeof(struct f2fs_bio_info),
+					GFP_KERNEL);
 		if (!sbi->write_io[i]) {
 			err = -ENOMEM;
 			goto free_options;
@@ -2517,14 +2604,14 @@ try_onemore:
 
 	err = init_percpu_info(sbi);
 	if (err)
-		goto free_options;
+		goto free_bio_info;
 
 	if (F2FS_IO_SIZE(sbi) > 1) {
 		sbi->write_io_dummy =
 			mempool_create_page_pool(2 * (F2FS_IO_SIZE(sbi) - 1), 0);
 		if (!sbi->write_io_dummy) {
 			err = -ENOMEM;
-			goto free_options;
+			goto free_percpu;
 		}
 	}
 
@@ -2559,6 +2646,7 @@ try_onemore:
 	sbi->last_valid_block_count = sbi->total_valid_block_count;
 	sbi->reserved_blocks = 0;
 	sbi->current_reserved_blocks = 0;
+	limit_reserve_root(sbi);
 
 	for (i = 0; i < NR_INODE_TYPE; i++) {
 		INIT_LIST_HEAD(&sbi->inode_list[i]);
@@ -2604,18 +2692,16 @@ try_onemore:
 		goto free_nm;
 	}
 
-	f2fs_join_shrinker(sbi);
-
 	err = f2fs_build_stats(sbi);
 	if (err)
-		goto free_nm;
+		goto free_node_inode;
 
 	/* read root inode and dentry */
 	root = f2fs_iget(sb, F2FS_ROOT_INO(sbi));
 	if (IS_ERR(root)) {
 		f2fs_msg(sb, KERN_ERR, "Failed to read root inode");
 		err = PTR_ERR(root);
-		goto free_node_inode;
+		goto free_stats;
 	}
 	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
 		iput(root);
@@ -2711,6 +2797,8 @@ skip_recovery:
 			sbi->valid_super_block ? 1 : 2, err);
 	}
 
+	f2fs_join_shrinker(sbi);
+
 	f2fs_msg(sbi->sb, KERN_NOTICE, "Mounted with checkpoint version = %llx",
 				cur_cp_version(F2FS_CKPT(sbi)));
 	f2fs_update_time(sbi, CP_TIME);
@@ -2737,14 +2825,12 @@ free_sysfs:
 free_root_inode:
 	dput(sb->s_root);
 	sb->s_root = NULL;
+free_stats:
+	f2fs_destroy_stats(sbi);
 free_node_inode:
-	truncate_inode_pages_final(NODE_MAPPING(sbi));
-	mutex_lock(&sbi->umount_mutex);
 	release_ino_entry(sbi, true);
-	f2fs_leave_shrinker(sbi);
+	truncate_inode_pages_final(NODE_MAPPING(sbi));
 	iput(sbi->node_inode);
-	mutex_unlock(&sbi->umount_mutex);
-	f2fs_destroy_stats(sbi);
 free_nm:
 	destroy_node_manager(sbi);
 free_sm:
@@ -2757,10 +2843,12 @@ free_meta_inode:
 	iput(sbi->meta_inode);
 free_io_dummy:
 	mempool_destroy(sbi->write_io_dummy);
-free_options:
+free_percpu:
+	destroy_percpu_info(sbi);
+free_bio_info:
 	for (i = 0; i < NR_PAGE_TYPE; i++)
 		kfree(sbi->write_io[i]);
-	destroy_percpu_info(sbi);
+free_options:
 #ifdef CONFIG_QUOTA
 	for (i = 0; i < MAXQUOTAS; i++)
 		kfree(sbi->s_qf_names[i]);

fs/f2fs/sysfs.c

@@ -113,6 +113,9 @@ static ssize_t features_show(struct f2fs_attr *a,
 	if (f2fs_sb_has_quota_ino(sb))
 		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "quota_ino");
+	if (f2fs_sb_has_inode_crtime(sb))
+		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "inode_crtime");
 	len += snprintf(buf + len, PAGE_SIZE - len, "\n");
 	return len;
 }
@@ -162,7 +165,8 @@ static ssize_t f2fs_sbi_store(struct f2fs_attr *a,
 #endif
 	if (a->struct_type == RESERVED_BLOCKS) {
 		spin_lock(&sbi->stat_lock);
-		if (t > (unsigned long)sbi->user_block_count) {
+		if (t > (unsigned long)(sbi->user_block_count -
+					sbi->root_reserved_blocks)) {
 			spin_unlock(&sbi->stat_lock);
 			return -EINVAL;
 		}
@@ -231,6 +235,7 @@ enum feat_id {
 	FEAT_INODE_CHECKSUM,
 	FEAT_FLEXIBLE_INLINE_XATTR,
 	FEAT_QUOTA_INO,
+	FEAT_INODE_CRTIME,
 };
 
 static ssize_t f2fs_feature_show(struct f2fs_attr *a,
@@ -245,6 +250,7 @@ static ssize_t f2fs_feature_show(struct f2fs_attr *a,
 	case FEAT_INODE_CHECKSUM:
 	case FEAT_FLEXIBLE_INLINE_XATTR:
 	case FEAT_QUOTA_INO:
+	case FEAT_INODE_CRTIME:
 		return snprintf(buf, PAGE_SIZE, "supported\n");
 	}
 	return 0;
@@ -299,6 +305,8 @@ F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, dir_level, dir_level);
 F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, cp_interval, interval_time[CP_TIME]);
 F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, idle_interval, interval_time[REQ_TIME]);
 F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, iostat_enable, iostat_enable);
+F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, readdir_ra, readdir_ra);
+F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, gc_pin_file_thresh, gc_pin_file_threshold);
 #ifdef CONFIG_F2FS_FAULT_INJECTION
 F2FS_RW_ATTR(FAULT_INFO_RATE, f2fs_fault_info, inject_rate, inject_rate);
 F2FS_RW_ATTR(FAULT_INFO_TYPE, f2fs_fault_info, inject_type, inject_type);
@@ -320,6 +328,7 @@ F2FS_FEATURE_RO_ATTR(project_quota, FEAT_PROJECT_QUOTA);
 F2FS_FEATURE_RO_ATTR(inode_checksum, FEAT_INODE_CHECKSUM);
 F2FS_FEATURE_RO_ATTR(flexible_inline_xattr, FEAT_FLEXIBLE_INLINE_XATTR);
 F2FS_FEATURE_RO_ATTR(quota_ino, FEAT_QUOTA_INO);
+F2FS_FEATURE_RO_ATTR(inode_crtime, FEAT_INODE_CRTIME);
 
 #define ATTR_LIST(name) (&f2fs_attr_##name.attr)
 static struct attribute *f2fs_attrs[] = {
@@ -346,6 +355,8 @@ static struct attribute *f2fs_attrs[] = {
 	ATTR_LIST(cp_interval),
 	ATTR_LIST(idle_interval),
 	ATTR_LIST(iostat_enable),
+	ATTR_LIST(readdir_ra),
+	ATTR_LIST(gc_pin_file_thresh),
 #ifdef CONFIG_F2FS_FAULT_INJECTION
 	ATTR_LIST(inject_rate),
 	ATTR_LIST(inject_type),
@@ -371,6 +382,7 @@ static struct attribute *f2fs_feat_attrs[] = {
 	ATTR_LIST(inode_checksum),
 	ATTR_LIST(flexible_inline_xattr),
 	ATTR_LIST(quota_ino),
+	ATTR_LIST(inode_crtime),
 	NULL,
 };
 

fs/f2fs/trace.c

@@ -17,7 +17,7 @@
 #include "trace.h"
 
 static RADIX_TREE(pids, GFP_ATOMIC);
-static spinlock_t pids_lock;
+static struct mutex pids_lock;
 static struct last_io_info last_io;
 
 static inline void __print_last_io(void)
@@ -64,7 +64,7 @@ void f2fs_trace_pid(struct page *page)
 	if (radix_tree_preload(GFP_NOFS))
 		return;
 
-	spin_lock(&pids_lock);
+	mutex_lock(&pids_lock);
 	p = radix_tree_lookup(&pids, pid);
 	if (p == current)
 		goto out;
@@ -77,7 +77,7 @@ void f2fs_trace_pid(struct page *page)
 			MAJOR(inode->i_sb->s_dev), MINOR(inode->i_sb->s_dev),
 			pid, current->comm);
 out:
-	spin_unlock(&pids_lock);
+	mutex_unlock(&pids_lock);
 	radix_tree_preload_end();
 }
 
@@ -122,7 +122,7 @@ void f2fs_trace_ios(struct f2fs_io_info *fio, int flush)
 
 void f2fs_build_trace_ios(void)
 {
-	spin_lock_init(&pids_lock);
+	mutex_init(&pids_lock);
 }
 
 #define PIDVEC_SIZE	128
@@ -150,7 +150,7 @@ void f2fs_destroy_trace_ios(void)
 	pid_t next_pid = 0;
 	unsigned int found;
 
-	spin_lock(&pids_lock);
+	mutex_lock(&pids_lock);
 	while ((found = gang_lookup_pids(pid, next_pid, PIDVEC_SIZE))) {
 		unsigned idx;
 
@@ -158,5 +158,5 @@ void f2fs_destroy_trace_ios(void)
 		for (idx = 0; idx < found; idx++)
 			radix_tree_delete(&pids, pid[idx]);
 	}
-	spin_unlock(&pids_lock);
+	mutex_unlock(&pids_lock);
 }

fs/f2fs/xattr.c

@@ -298,8 +298,8 @@ static int lookup_all_xattrs(struct inode *inode, struct page *ipage,
 	if (!size && !inline_size)
 		return -ENODATA;
 
-	txattr_addr = kzalloc(inline_size + size + XATTR_PADDING_SIZE,
-							GFP_F2FS_ZERO);
+	txattr_addr = f2fs_kzalloc(F2FS_I_SB(inode),
+			inline_size + size + XATTR_PADDING_SIZE, GFP_NOFS);
 	if (!txattr_addr)
 		return -ENOMEM;
 
@@ -351,8 +351,8 @@ static int read_all_xattrs(struct inode *inode, struct page *ipage,
 	void *txattr_addr;
 	int err;
 
-	txattr_addr = kzalloc(inline_size + size + XATTR_PADDING_SIZE,
-							GFP_F2FS_ZERO);
+	txattr_addr = f2fs_kzalloc(F2FS_I_SB(inode),
+			inline_size + size + XATTR_PADDING_SIZE, GFP_NOFS);
 	if (!txattr_addr)
 		return -ENOMEM;
 
@@ -433,6 +433,7 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 	if (F2FS_I(inode)->i_xattr_nid) {
 		xpage = get_node_page(sbi, F2FS_I(inode)->i_xattr_nid);
 		if (IS_ERR(xpage)) {
+			err = PTR_ERR(xpage);
 			alloc_nid_failed(sbi, new_nid);
 			goto in_page_out;
 		}
@@ -443,6 +444,7 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 		set_new_dnode(&dn, inode, NULL, NULL, new_nid);
 		xpage = new_node_page(&dn, XATTR_NODE_OFFSET);
 		if (IS_ERR(xpage)) {
+			err = PTR_ERR(xpage);
 			alloc_nid_failed(sbi, new_nid);
 			goto in_page_out;
 		}
@@ -598,7 +600,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 			goto exit;
 		}
 
-		if (f2fs_xattr_value_same(here, value, size))
+		if (value && f2fs_xattr_value_same(here, value, size))
 			goto exit;
 	} else if ((flags & XATTR_REPLACE)) {
 		error = -ENODATA;

fs/posix_acl.c

@@ -43,7 +43,7 @@ struct posix_acl *get_cached_acl(struct inode *inode, int type)
 		rcu_read_lock();
 		acl = rcu_dereference(*p);
 		if (!acl || is_uncached_acl(acl) ||
-		    atomic_inc_not_zero(&acl->a_refcount))
+		    refcount_inc_not_zero(&acl->a_refcount))
 			break;
 		rcu_read_unlock();
 		cpu_relax();
@@ -164,7 +164,7 @@ EXPORT_SYMBOL(get_acl);
 void
 posix_acl_init(struct posix_acl *acl, int count)
 {
-	atomic_set(&acl->a_refcount, 1);
+	refcount_set(&acl->a_refcount, 1);
 	acl->a_count = count;
 }
 EXPORT_SYMBOL(posix_acl_init);
@@ -197,7 +197,7 @@ posix_acl_clone(const struct posix_acl *acl, gfp_t flags)
 		           sizeof(struct posix_acl_entry);
 		clone = kmemdup(acl, size, flags);
 		if (clone)
-			atomic_set(&clone->a_refcount, 1);
+			refcount_set(&clone->a_refcount, 1);
 	}
 	return clone;
 }

include/linux/f2fs_fs.h

@@ -117,6 +117,7 @@ struct f2fs_super_block {
 /*
  * For checkpoint
  */
+#define CP_NOCRC_RECOVERY_FLAG	0x00000200
 #define CP_TRIMMED_FLAG		0x00000100
 #define CP_NAT_BITS_FLAG	0x00000080
 #define CP_CRC_RECOVERY_FLAG	0x00000040
@@ -212,6 +213,7 @@ struct f2fs_extent {
 #define F2FS_DATA_EXIST		0x08	/* file inline data exist flag */
 #define F2FS_INLINE_DOTS	0x10	/* file having implicit dot dentries */
 #define F2FS_EXTRA_ATTR		0x20	/* file having extra attribute */
+#define F2FS_PIN_FILE		0x40	/* file should not be gced */
 
 struct f2fs_inode {
 	__le16 i_mode;			/* file mode */
@@ -229,7 +231,13 @@ struct f2fs_inode {
 	__le32 i_ctime_nsec;		/* change time in nano scale */
 	__le32 i_mtime_nsec;		/* modification time in nano scale */
 	__le32 i_generation;		/* file version (for NFS) */
-	__le32 i_current_depth;		/* only for directory depth */
+	union {
+		__le32 i_current_depth;	/* only for directory depth */
+		__le16 i_gc_failures;	/*
+					 * # of gc failures on pinned file.
+					 * only for regular files.
+					 */
+	};
 	__le32 i_xattr_nid;		/* nid to save xattr */
 	__le32 i_flags;			/* file attributes */
 	__le32 i_pino;			/* parent inode number */
@@ -245,8 +253,10 @@ struct f2fs_inode {
 			__le16 i_inline_xattr_size;	/* inline xattr size, unit: 4 bytes */
 			__le32 i_projid;	/* project id */
 			__le32 i_inode_checksum;/* inode meta checksum */
+			__le64 i_crtime;	/* creation time */
+			__le32 i_crtime_nsec;	/* creation time in nano scale */
 			__le32 i_extra_end[0];	/* for attribute size calculation */
-		};
+		} __packed;
 		__le32 i_addr[DEF_ADDRS_PER_INODE];	/* Pointers to data blocks */
 	};
 	__le32 i_nid[DEF_NIDS_PER_INODE];	/* direct(2), indirect(2),

include/linux/posix_acl.h

@@ -12,6 +12,7 @@
 #include <linux/bug.h>
 #include <linux/slab.h>
 #include <linux/rcupdate.h>
+#include <linux/refcount.h>
 #include <uapi/linux/posix_acl.h>
 
 struct posix_acl_entry {
@@ -24,7 +25,7 @@ struct posix_acl_entry {
 };
 
 struct posix_acl {
-	atomic_t		a_refcount;
+	refcount_t		a_refcount;
 	struct rcu_head		a_rcu;
 	unsigned int		a_count;
 	struct posix_acl_entry	a_entries[0];
@@ -41,7 +42,7 @@ static inline struct posix_acl *
 posix_acl_dup(struct posix_acl *acl)
 {
 	if (acl)
-		atomic_inc(&acl->a_refcount);
+		refcount_inc(&acl->a_refcount);
 	return acl;
 }
 
@@ -51,7 +52,7 @@ posix_acl_dup(struct posix_acl *acl)
 static inline void
 posix_acl_release(struct posix_acl *acl)
 {
-	if (acl && atomic_dec_and_test(&acl->a_refcount))
+	if (acl && refcount_dec_and_test(&acl->a_refcount))
 		kfree_rcu(acl, a_rcu);
 }
 

include/trace/events/f2fs.h

@@ -147,7 +147,8 @@ TRACE_DEFINE_ENUM(CP_TRIMMED);
 		{ CP_NO_SPC_ROLL,	"no space roll forward" },	\
 		{ CP_NODE_NEED_CP,	"node needs cp" },		\
 		{ CP_FASTBOOT_MODE,	"fastboot mode" },		\
-		{ CP_SPEC_LOG_NUM,	"log type is 2" })
+		{ CP_SPEC_LOG_NUM,	"log type is 2" },		\
+		{ CP_RECOVER_DIR,	"dir needs recovery" })
 
 struct victim_sel_policy;
 struct f2fs_map_blocks;