Merge tag 'f2fs-for-6.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
"This series introduces a device aliasing feature where user can carve
out partitions but reclaim the space back by deleting aliased file in
root dir.

In addition to that, there're numerous minor bug fixes in zoned device
support, checkpoint=disable, extent cache management, fiemap, and
lazytime mount option. The full list of noticeable changes can be
found below.

Enhancements:
- introduce device aliasing file
- add stats in debugfs to show multiple devices
- add a sysfs node to limit max read extent count per-inode
- modify f2fs_is_checkpoint_ready logic to allow more data to be
written with the CP disable
- decrease spare area for pinned files for zoned devices

Fixes:
- Revert "f2fs: remove unreachable lazytime mount option parsing"
- adjust unusable cap before checkpoint=disable mode
- fix to drop all discards after creating snapshot on lvm device
- fix to shrink read extent node in batches
- fix changing cursegs if recovery fails on zoned device
- fix to adjust appropriate length for fiemap
- fix fiemap failure issue when page size is 16KB
- fix to avoid forcing direct write to use buffered IO on inline_data
inode
- fix to map blocks correctly for direct write
- fix to account dirty data in __get_secs_required()
- fix null-ptr-deref in f2fs_submit_page_bio()
- fix inconsistent update of i_blocks in release_compress_blocks and
reserve_compress_blocks"

* tag 'f2fs-for-6.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (40 commits)
f2fs: fix to drop all discards after creating snapshot on lvm device
f2fs: add a sysfs node to limit max read extent count per-inode
f2fs: fix to shrink read extent node in batches
f2fs: print message if fscorrupted was found in f2fs_new_node_page()
f2fs: clear SBI_POR_DOING before initing inmem curseg
f2fs: fix changing cursegs if recovery fails on zoned device
f2fs: adjust unusable cap before checkpoint=disable mode
f2fs: fix to requery extent which cross boundary of inquiry
f2fs: fix to adjust appropriate length for fiemap
f2fs: clean up w/ F2FS_{BLK_TO_BYTES,BTYES_TO_BLK}
f2fs: fix to do cast in F2FS_{BLK_TO_BYTES, BTYES_TO_BLK} to avoid overflow
f2fs: replace deprecated strcpy with strscpy
Revert "f2fs: remove unreachable lazytime mount option parsing"
f2fs: fix to avoid forcing direct write to use buffered IO on inline_data inode
f2fs: fix to map blocks correctly for direct write
f2fs: fix race in concurrent f2fs_stop_gc_thread
f2fs: fix fiemap failure issue when page size is 16KB
f2fs: remove redundant atomic file check in defragment
f2fs: fix to convert log type to segment data type correctly
f2fs: clean up the unused variable additional_reserved_segments
...

Documentation/ABI/testing/sysfs-fs-f2fs

···
 		GC approach and turns SSR mode on.
 		gc urgent low(2): lowers the bar of checking I/O idling in
 		order to process outstanding discard commands and GC a
-		little bit aggressively. uses cost benefit GC approach.
+		little bit aggressively. always uses cost benefit GC approach,
+		and will override age-threshold GC approach if ATGC is enabled
+		at the same time.
 		gc urgent mid(3): does GC forcibly in a period of given
 		gc_urgent_sleep_time and executes a mid level of I/O idling check.
-		uses cost benefit GC approach.
+		always uses cost benefit GC approach, and will override
+		age-threshold GC approach if ATGC is enabled at the same time.
 
 What:		/sys/fs/f2fs/<disk>/gc_urgent_sleep_time
 Date:		August 2017
···
 		for zoned deivces. The initial value of it is 95(%). F2FS will stop the
 		background GC thread from intiating GC for sections having valid blocks
 		exceeding the ratio.
+
+What:		/sys/fs/f2fs/<disk>/max_read_extent_count
+Date:		November 2024
+Contact:	"Chao Yu" <[e-mail]>
+Description:	It controls max read extent count for per-inode, the value of threshold
+		is 10240 by default.

Documentation/filesystems/f2fs.rst

···
   can start before the zone-capacity and span across zone-capacity boundary.
   Such spanning segments are also considered as usable segments. All blocks
   past the zone-capacity are considered unusable in these segments.
+
+Device aliasing feature
+-----------------------
+
+f2fs can utilize a special file called a "device aliasing file." This file allows
+the entire storage device to be mapped with a single, large extent, not using
+the usual f2fs node structures. This mapped area is pinned and primarily intended
+for holding the space.
+
+Essentially, this mechanism allows a portion of the f2fs area to be temporarily
+reserved and used by another filesystem or for different purposes. Once that
+external usage is complete, the device aliasing file can be deleted, releasing
+the reserved space back to F2FS for its own use.
+
+<use-case>
+
+# ls /dev/vd*
+/dev/vdb (32GB) /dev/vdc (32GB)
+# mkfs.ext4 /dev/vdc
+# mkfs.f2fs -c /dev/[e-mail] /dev/vdb
+# mount /dev/vdb /mnt/f2fs
+# ls -l /mnt/f2fs
+vdc.file
+# df -h
+/dev/vdb                            64G   33G   32G  52% /mnt/f2fs
+
+# mount -o loop /dev/vdc /mnt/ext4
+# df -h
+/dev/vdb                            64G   33G   32G  52% /mnt/f2fs
+/dev/loop7                          32G   24K   30G   1% /mnt/ext4
+# umount /mnt/ext4
+
+# f2fs_io getflags /mnt/f2fs/vdc.file
+get a flag on /mnt/f2fs/vdc.file ret=0, flags=nocow(pinned),immutable
+# f2fs_io setflags noimmutable /mnt/f2fs/vdc.file
+get a flag on noimmutable ret=0, flags=800010
+set a flag on /mnt/f2fs/vdc.file ret=0, flags=noimmutable
+# rm /mnt/f2fs/vdc.file
+# df -h
+/dev/vdb                            64G  753M   64G   2% /mnt/f2fs
+
+So, the key idea is, user can do any file operations on /dev/vdc, and
+reclaim the space after the use, while the space is counted as /data.
+That doesn't require modifying partition size and filesystem format.

fs/f2fs/acl.c

···
 	struct posix_acl *clone = NULL;
 
 	if (acl) {
-		int size = sizeof(struct posix_acl) + acl->a_count *
-				sizeof(struct posix_acl_entry);
-		clone = kmemdup(acl, size, flags);
+		clone = kmemdup(acl, struct_size(acl, a_entries, acl->a_count),
+				flags);
 		if (clone)
 			refcount_set(&clone->a_refcount, 1);
 	}

fs/f2fs/checkpoint.c

···
 	f2fs_build_fault_attr(sbi, 0, 0);
 	if (!end_io)
 		f2fs_flush_merged_writes(sbi);
-	f2fs_handle_critical_error(sbi, reason, end_io);
+	f2fs_handle_critical_error(sbi, reason);
 }
 
 /*

fs/f2fs/data.c

···
 		/* reserved delalloc block should be mapped for fiemap. */
 		if (blkaddr == NEW_ADDR)
 			map->m_flags |= F2FS_MAP_DELALLOC;
-		if (flag != F2FS_GET_BLOCK_DIO || !is_hole)
+		/* DIO READ and hole case, should not map the blocks. */
+		if (!(flag == F2FS_GET_BLOCK_DIO && is_hole && !map->m_may_create))
 			map->m_flags |= F2FS_MAP_MAPPED;
 
 		map->m_pblk = blkaddr;
···
 	return true;
 }
 
-static inline u64 bytes_to_blks(struct inode *inode, u64 bytes)
-{
-	return (bytes >> inode->i_blkbits);
-}
-
-static inline u64 blks_to_bytes(struct inode *inode, u64 blks)
-{
-	return (blks << inode->i_blkbits);
-}
-
 static int f2fs_xattr_fiemap(struct inode *inode,
 				struct fiemap_extent_info *fieinfo)
 {
···
 			return err;
 		}
 
-		phys = blks_to_bytes(inode, ni.blk_addr);
+		phys = F2FS_BLK_TO_BYTES(ni.blk_addr);
 		offset = offsetof(struct f2fs_inode, i_addr) +
 					sizeof(__le32) * (DEF_ADDRS_PER_INODE -
 					get_inline_xattr_addrs(inode));
···
 			return err;
 		}
 
-		phys = blks_to_bytes(inode, ni.blk_addr);
+		phys = F2FS_BLK_TO_BYTES(ni.blk_addr);
 		len = inode->i_sb->s_blocksize;
 
 		f2fs_put_page(page, 1);
···
 	return (err < 0 ? err : 0);
 }
 
-static loff_t max_inode_blocks(struct inode *inode)
-{
-	loff_t result = ADDRS_PER_INODE(inode);
-	loff_t leaf_count = ADDRS_PER_BLOCK(inode);
-
-	/* two direct node blocks */
-	result += (leaf_count * 2);
-
-	/* two indirect node blocks */
-	leaf_count *= NIDS_PER_BLOCK;
-	result += (leaf_count * 2);
-
-	/* one double indirect node block */
-	leaf_count *= NIDS_PER_BLOCK;
-	result += leaf_count;
-
-	return result;
-}
-
 int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		u64 start, u64 len)
 {
 	struct f2fs_map_blocks map;
-	sector_t start_blk, last_blk;
+	sector_t start_blk, last_blk, blk_len, max_len;
 	pgoff_t next_pgofs;
 	u64 logical = 0, phys = 0, size = 0;
 	u32 flags = 0;
···
 			goto out;
 	}
 
-	if (bytes_to_blks(inode, len) == 0)
-		len = blks_to_bytes(inode, 1);
-
-	start_blk = bytes_to_blks(inode, start);
-	last_blk = bytes_to_blks(inode, start + len - 1);
+	start_blk = F2FS_BYTES_TO_BLK(start);
+	last_blk = F2FS_BYTES_TO_BLK(start + len - 1);
+	blk_len = last_blk - start_blk + 1;
+	max_len = F2FS_BYTES_TO_BLK(maxbytes) - start_blk;
 
 next:
 	memset(&map, 0, sizeof(map));
 	map.m_lblk = start_blk;
-	map.m_len = bytes_to_blks(inode, len);
+	map.m_len = blk_len;
 	map.m_next_pgofs = &next_pgofs;
 	map.m_seg_type = NO_CHECK_TYPE;
 
···
 	if (!compr_cluster && !(map.m_flags & F2FS_MAP_FLAGS)) {
 		start_blk = next_pgofs;
 
-		if (blks_to_bytes(inode, start_blk) < blks_to_bytes(inode,
-						max_inode_blocks(inode)))
+		if (F2FS_BLK_TO_BYTES(start_blk) < maxbytes)
 			goto prep_next;
 
 		flags |= FIEMAP_EXTENT_LAST;
+	}
+
+	/*
+	 * current extent may cross boundary of inquiry, increase len to
+	 * requery.
+	 */
+	if (!compr_cluster && (map.m_flags & F2FS_MAP_MAPPED) &&
+				map.m_lblk + map.m_len - 1 == last_blk &&
+				blk_len != max_len) {
+		blk_len = max_len;
+		goto next;
 	}
 
 	compr_appended = false;
···
 	} else if (compr_appended) {
 		unsigned int appended_blks = cluster_size -
 						count_in_cluster + 1;
-		size += blks_to_bytes(inode, appended_blks);
+		size += F2FS_BLK_TO_BYTES(appended_blks);
 		start_blk += appended_blks;
 		compr_cluster = false;
 	} else {
-		logical = blks_to_bytes(inode, start_blk);
+		logical = F2FS_BLK_TO_BYTES(start_blk);
 		phys = __is_valid_data_blkaddr(map.m_pblk) ?
-			blks_to_bytes(inode, map.m_pblk) : 0;
-		size = blks_to_bytes(inode, map.m_len);
+			F2FS_BLK_TO_BYTES(map.m_pblk) : 0;
+		size = F2FS_BLK_TO_BYTES(map.m_len);
 		flags = 0;
 
 		if (compr_cluster) {
···
 			count_in_cluster += map.m_len;
 			if (count_in_cluster == cluster_size) {
 				compr_cluster = false;
-				size += blks_to_bytes(inode, 1);
+				size += F2FS_BLKSIZE;
 			}
 		} else if (map.m_flags & F2FS_MAP_DELALLOC) {
 			flags = FIEMAP_EXTENT_UNWRITTEN;
 		}
 
-		start_blk += bytes_to_blks(inode, size);
+		start_blk += F2FS_BYTES_TO_BLK(size);
 	}
 
 prep_next:
···
 					struct readahead_control *rac)
 {
 	struct bio *bio = *bio_ret;
-	const unsigned blocksize = blks_to_bytes(inode, 1);
+	const unsigned int blocksize = F2FS_BLKSIZE;
 	sector_t block_in_file;
 	sector_t last_block;
 	sector_t last_block_in_file;
···
 
 	block_in_file = (sector_t)index;
 	last_block = block_in_file + nr_pages;
-	last_block_in_file = bytes_to_blks(inode,
-			f2fs_readpage_limit(inode) + blocksize - 1);
+	last_block_in_file = F2FS_BYTES_TO_BLK(f2fs_readpage_limit(inode) +
+							blocksize - 1);
 	if (last_block > last_block_in_file)
 		last_block = last_block_in_file;
 
···
 	struct bio *bio = *bio_ret;
 	unsigned int start_idx = cc->cluster_idx << cc->log_cluster_size;
 	sector_t last_block_in_file;
-	const unsigned blocksize = blks_to_bytes(inode, 1);
+	const unsigned int blocksize = F2FS_BLKSIZE;
 	struct decompress_io_ctx *dic = NULL;
 	struct extent_info ei = {};
 	bool from_dnode = true;
···
 
 	f2fs_bug_on(sbi, f2fs_cluster_is_empty(cc));
 
-	last_block_in_file = bytes_to_blks(inode,
-			f2fs_readpage_limit(inode) + blocksize - 1);
+	last_block_in_file = F2FS_BYTES_TO_BLK(f2fs_readpage_limit(inode) +
+							blocksize - 1);
 
 	/* get rid of pages beyond EOF */
 	for (i = 0; i < cc->cluster_size; i++) {
···
 		.nr_cpages = 0,
 	};
 	pgoff_t nc_cluster_idx = NULL_CLUSTER;
+	pgoff_t index;
 #endif
 	unsigned nr_pages = rac ? readahead_count(rac) : 1;
 	unsigned max_nr_pages = nr_pages;
-	pgoff_t index;
 	int ret = 0;
 
 	map.m_pblk = 0;
···
 			prefetchw(&folio->flags);
 		}
 
+#ifdef CONFIG_F2FS_FS_COMPRESSION
 		index = folio_index(folio);
 
-#ifdef CONFIG_F2FS_FS_COMPRESSION
 		if (!f2fs_compressed_file(inode))
 			goto read_single_page;
 
···
 
 	if (!f2fs_lookup_read_extent_cache_block(inode, index,
 						 &dn.data_blkaddr)) {
+		if (IS_DEVICE_ALIASING(inode)) {
+			err = -ENODATA;
+			goto out;
+		}
+
 		if (locked) {
 			err = f2fs_reserve_block(&dn, index);
 			goto out;
···
 	 * to be very smart.
 	 */
 	cur_lblock = 0;
-	last_lblock = bytes_to_blks(inode, i_size_read(inode));
+	last_lblock = F2FS_BYTES_TO_BLK(i_size_read(inode));
 
 	while (cur_lblock < last_lblock && cur_lblock < sis->max) {
 		struct f2fs_map_blocks map;
···
 	pgoff_t next_pgofs = 0;
 	int err;
 
-	map.m_lblk = bytes_to_blks(inode, offset);
-	map.m_len = bytes_to_blks(inode, offset + length - 1) - map.m_lblk + 1;
+	map.m_lblk = F2FS_BYTES_TO_BLK(offset);
+	map.m_len = F2FS_BYTES_TO_BLK(offset + length - 1) - map.m_lblk + 1;
 	map.m_next_pgofs = &next_pgofs;
 	map.m_seg_type = f2fs_rw_hint_to_seg_type(F2FS_I_SB(inode),
 						inode->i_write_hint);
···
 	if (err)
 		return err;
 
-	iomap->offset = blks_to_bytes(inode, map.m_lblk);
+	iomap->offset = F2FS_BLK_TO_BYTES(map.m_lblk);
 
 	/*
 	 * When inline encryption is enabled, sometimes I/O to an encrypted file
···
 		if (WARN_ON_ONCE(map.m_pblk == NEW_ADDR))
 			return -EINVAL;
 
-		iomap->length = blks_to_bytes(inode, map.m_len);
+		iomap->length = F2FS_BLK_TO_BYTES(map.m_len);
 		iomap->type = IOMAP_MAPPED;
 		iomap->flags |= IOMAP_F_MERGED;
 		iomap->bdev = map.m_bdev;
-		iomap->addr = blks_to_bytes(inode, map.m_pblk);
+		iomap->addr = F2FS_BLK_TO_BYTES(map.m_pblk);
 	} else {
 		if (flags & IOMAP_WRITE)
 			return -ENOTBLK;
 
 		if (map.m_pblk == NULL_ADDR) {
-			iomap->length = blks_to_bytes(inode, next_pgofs) -
-								iomap->offset;
+			iomap->length = F2FS_BLK_TO_BYTES(next_pgofs) -
+							iomap->offset;
 			iomap->type = IOMAP_HOLE;
 		} else if (map.m_pblk == NEW_ADDR) {
-			iomap->length = blks_to_bytes(inode, map.m_len);
+			iomap->length = F2FS_BLK_TO_BYTES(map.m_len);
 			iomap->type = IOMAP_UNWRITTEN;
 		} else {
 			f2fs_bug_on(F2FS_I_SB(inode), 1);

fs/f2fs/debug.c

···
 }
 
 #ifdef CONFIG_DEBUG_FS
+static void update_multidevice_stats(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_stat_info *si = F2FS_STAT(sbi);
+	struct f2fs_dev_stats *dev_stats = si->dev_stats;
+	int i, j;
+
+	if (!f2fs_is_multi_device(sbi))
+		return;
+
+	memset(dev_stats, 0, sizeof(struct f2fs_dev_stats) * sbi->s_ndevs);
+	for (i = 0; i < sbi->s_ndevs; i++) {
+		unsigned int start_segno, end_segno;
+		block_t start_blk, end_blk;
+
+		if (i == 0) {
+			start_blk = MAIN_BLKADDR(sbi);
+			end_blk = FDEV(i).end_blk + 1 - SEG0_BLKADDR(sbi);
+		} else {
+			start_blk = FDEV(i).start_blk;
+			end_blk = FDEV(i).end_blk + 1;
+		}
+
+		start_segno = GET_SEGNO(sbi, start_blk);
+		end_segno = GET_SEGNO(sbi, end_blk);
+
+		for (j = start_segno; j < end_segno; j++) {
+			unsigned int seg_blks, sec_blks;
+
+			seg_blks = get_seg_entry(sbi, j)->valid_blocks;
+
+			/* update segment stats */
+			if (IS_CURSEG(sbi, j))
+				dev_stats[i].devstats[0][DEVSTAT_INUSE]++;
+			else if (seg_blks == BLKS_PER_SEG(sbi))
+				dev_stats[i].devstats[0][DEVSTAT_FULL]++;
+			else if (seg_blks != 0)
+				dev_stats[i].devstats[0][DEVSTAT_DIRTY]++;
+			else if (!test_bit(j, FREE_I(sbi)->free_segmap))
+				dev_stats[i].devstats[0][DEVSTAT_FREE]++;
+			else
+				dev_stats[i].devstats[0][DEVSTAT_PREFREE]++;
+
+			if (!__is_large_section(sbi) ||
+				(j % SEGS_PER_SEC(sbi)) != 0)
+				continue;
+
+			sec_blks = get_sec_entry(sbi, j)->valid_blocks;
+
+			/* update section stats */
+			if (IS_CURSEC(sbi, GET_SEC_FROM_SEG(sbi, j)))
+				dev_stats[i].devstats[1][DEVSTAT_INUSE]++;
+			else if (sec_blks == BLKS_PER_SEC(sbi))
+				dev_stats[i].devstats[1][DEVSTAT_FULL]++;
+			else if (sec_blks != 0)
+				dev_stats[i].devstats[1][DEVSTAT_DIRTY]++;
+			else if (!test_bit(GET_SEC_FROM_SEG(sbi, j),
+					FREE_I(sbi)->free_secmap))
+				dev_stats[i].devstats[1][DEVSTAT_FREE]++;
+			else
+				dev_stats[i].devstats[1][DEVSTAT_PREFREE]++;
+		}
+	}
+}
+
 static void update_general_status(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_stat_info *si = F2FS_STAT(sbi);
···
 			si->dirty_seg[type]++;
 		si->valid_blks[type] += blks;
 	}
+
+	update_multidevice_stats(sbi);
 
 	for (i = 0; i < MAX_CALL_TYPE; i++)
 		si->cp_call_count[i] = atomic_read(&sbi->cp_call_count[i]);
···
 			   si->dirty_count);
 		seq_printf(s, "  - Prefree: %d\n  - Free: %d (%d)\n\n",
 			   si->prefree_count, si->free_segs, si->free_secs);
+		if (f2fs_is_multi_device(sbi)) {
+			seq_puts(s, "Multidevice stats:\n");
+			seq_printf(s, "  [seg:   %8s %8s %8s %8s %8s]",
+					"inuse", "dirty", "full", "free", "prefree");
+			if (__is_large_section(sbi))
+				seq_printf(s, " [sec:   %8s %8s %8s %8s %8s]\n",
+					"inuse", "dirty", "full", "free", "prefree");
+			else
+				seq_puts(s, "\n");
+
+			for (i = 0; i < sbi->s_ndevs; i++) {
+				seq_printf(s, "  #%-2d     %8u %8u %8u %8u %8u", i,
+					si->dev_stats[i].devstats[0][DEVSTAT_INUSE],
+					si->dev_stats[i].devstats[0][DEVSTAT_DIRTY],
+					si->dev_stats[i].devstats[0][DEVSTAT_FULL],
+					si->dev_stats[i].devstats[0][DEVSTAT_FREE],
+					si->dev_stats[i].devstats[0][DEVSTAT_PREFREE]);
+				if (!__is_large_section(sbi)) {
+					seq_puts(s, "\n");
+					continue;
+				}
+				seq_printf(s, "          %8u %8u %8u %8u %8u\n",
+					si->dev_stats[i].devstats[1][DEVSTAT_INUSE],
+					si->dev_stats[i].devstats[1][DEVSTAT_DIRTY],
+					si->dev_stats[i].devstats[1][DEVSTAT_FULL],
+					si->dev_stats[i].devstats[1][DEVSTAT_FREE],
+					si->dev_stats[i].devstats[1][DEVSTAT_PREFREE]);
+			}
+			seq_puts(s, "\n");
+		}
 		seq_printf(s, "CP calls: %d (BG: %d)\n",
 			   si->cp_call_count[TOTAL_CALL],
 			   si->cp_call_count[BACKGROUND]);
···
 			   si->ndirty_node, si->node_pages);
 		seq_printf(s, "  - dents: %4d in dirs:%4d (%4d)\n",
 			   si->ndirty_dent, si->ndirty_dirs, si->ndirty_all);
-		seq_printf(s, "  - datas: %4d in files:%4d\n",
+		seq_printf(s, "  - data: %4d in files:%4d\n",
 			   si->ndirty_data, si->ndirty_files);
-		seq_printf(s, "  - quota datas: %4d in quota files:%4d\n",
+		seq_printf(s, "  - quota data: %4d in quota files:%4d\n",
 			   si->ndirty_qdata, si->nquota_files);
 		seq_printf(s, "  - meta: %4d in %4d\n",
 			   si->ndirty_meta, si->meta_pages);
···
 {
 	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
 	struct f2fs_stat_info *si;
+	struct f2fs_dev_stats *dev_stats;
 	unsigned long flags;
 	int i;
 
 	si = f2fs_kzalloc(sbi, sizeof(struct f2fs_stat_info), GFP_KERNEL);
 	if (!si)
 		return -ENOMEM;
+
+	dev_stats = f2fs_kzalloc(sbi, sizeof(struct f2fs_dev_stats) *
+						sbi->s_ndevs, GFP_KERNEL);
+	if (!dev_stats) {
+		kfree(si);
+		return -ENOMEM;
+	}
+
+	si->dev_stats = dev_stats;
 
 	si->all_area_segs = le32_to_cpu(raw_super->segment_count);
 	si->sit_area_segs = le32_to_cpu(raw_super->segment_count_sit);
···
 	list_del(&si->stat_list);
 	raw_spin_unlock_irqrestore(&f2fs_stat_lock, flags);
 
+	kfree(si->dev_stats);
 	kfree(si);
 }
 

fs/f2fs/extent_cache.c

···
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct f2fs_extent *i_ext = &F2FS_INODE(ipage)->i_ext;
 	struct extent_info ei;
+	int devi;
 
 	get_read_extent_info(&ei, i_ext);
 
···
 			  ei.blk, ei.fofs, ei.len);
 		return false;
 	}
-	return true;
+
+	if (!IS_DEVICE_ALIASING(inode))
+		return true;
+
+	for (devi = 0; devi < sbi->s_ndevs; devi++) {
+		if (FDEV(devi).start_blk != ei.blk ||
+				FDEV(devi).end_blk != ei.blk + ei.len - 1)
+			continue;
+
+		if (devi == 0) {
+			f2fs_warn(sbi,
+			    "%s: inode (ino=%lx) is an alias of meta device",
+			    __func__, inode->i_ino);
+			return false;
+		}
+
+		if (bdev_is_zoned(FDEV(devi).bdev)) {
+			f2fs_warn(sbi,
+			    "%s: device alias inode (ino=%lx)'s extent info "
+			    "[%u, %u, %u] maps to zoned block device",
+			    __func__, inode->i_ino, ei.blk, ei.fofs, ei.len);
+			return false;
+		}
+		return true;
+	}
+
+	f2fs_warn(sbi, "%s: device alias inode (ino=%lx)'s extent info "
+			"[%u, %u, %u] is inconsistent w/ any devices",
+			__func__, inode->i_ino, ei.blk, ei.fofs, ei.len);
+	return false;
 }
 
 static void __set_extent_info(struct extent_info *ei,
···
 
 static bool __may_extent_tree(struct inode *inode, enum extent_type type)
 {
+	if (IS_DEVICE_ALIASING(inode) && type == EX_READ)
+		return true;
+
 	/*
 	 * for recovered files during mount do not create extents
 	 * if shrinker is not registered.
···
 }
 
 static unsigned int __free_extent_tree(struct f2fs_sb_info *sbi,
-					struct extent_tree *et)
+				struct extent_tree *et, unsigned int nr_shrink)
 {
 	struct rb_node *node, *next;
 	struct extent_node *en;
-	unsigned int count = atomic_read(&et->node_cnt);
+	unsigned int count;
 
 	node = rb_first_cached(&et->root);
-	while (node) {
+
+	for (count = 0; node && count < nr_shrink; count++) {
 		next = rb_next(node);
 		en = rb_entry(node, struct extent_node, rb_node);
 		__release_extent_node(sbi, et, en);
 		node = next;
 	}
 
-	return count - atomic_read(&et->node_cnt);
+	return count;
 }
 
 static void __drop_largest_extent(struct extent_tree *et,
···
 	write_lock(&et->lock);
 	if (atomic_read(&et->node_cnt) || !ei.len)
 		goto skip;
+
+	if (IS_DEVICE_ALIASING(inode)) {
+		et->largest = ei;
+		goto skip;
+	}
 
 	en = __attach_extent_node(sbi, et, &ei, NULL,
 				&et->root.rb_root.rb_node, true);
···
 		*ei = et->largest;
 		ret = true;
 		stat_inc_largest_node_hit(sbi);
+		goto out;
+	}
+
+	if (IS_DEVICE_ALIASING(inode)) {
+		ret = false;
 		goto out;
 	}
 
···
 	return en;
 }
 
+static unsigned int __destroy_extent_node(struct inode *inode,
+					enum extent_type type)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct extent_tree *et = F2FS_I(inode)->extent_tree[type];
+	unsigned int nr_shrink = type == EX_READ ?
+				READ_EXTENT_CACHE_SHRINK_NUMBER :
+				AGE_EXTENT_CACHE_SHRINK_NUMBER;
+	unsigned int node_cnt = 0;
+
+	if (!et || !atomic_read(&et->node_cnt))
+		return 0;
+
+	while (atomic_read(&et->node_cnt)) {
+		write_lock(&et->lock);
+		node_cnt += __free_extent_tree(sbi, et, nr_shrink);
+		write_unlock(&et->lock);
+	}
+
+	f2fs_bug_on(sbi, atomic_read(&et->node_cnt));
+
+	return node_cnt;
+}
+
 static void __update_extent_tree_range(struct inode *inode,
 			struct extent_info *tei, enum extent_type type)
 {
···
 		}
 
 		if (end < org_end && (type != EX_READ ||
-				org_end - end >= F2FS_MIN_EXTENT_LEN)) {
+			(org_end - end >= F2FS_MIN_EXTENT_LEN &&
+			atomic_read(&et->node_cnt) <
+					sbi->max_read_extent_count))) {
 			if (parts) {
 				__set_extent_info(&ei,
 					end, org_end - end,
···
 		}
 	}
 
-	if (is_inode_flag_set(inode, FI_NO_EXTENT))
-		__free_extent_tree(sbi, et);
-
 	if (et->largest_updated) {
 		et->largest_updated = false;
 		updated = true;
···
 					insert_p, insert_parent, leftmost);
 out_read_extent_cache:
 	write_unlock(&et->lock);
+
+	if (is_inode_flag_set(inode, FI_NO_EXTENT))
+		__destroy_extent_node(inode, EX_READ);
 
 	if (updated)
 		f2fs_mark_inode_dirty_sync(inode, true);
···
 	list_for_each_entry_safe(et, next, &eti->zombie_list, list) {
 		if (atomic_read(&et->node_cnt)) {
 			write_lock(&et->lock);
-			node_cnt += __free_extent_tree(sbi, et);
+			node_cnt += __free_extent_tree(sbi, et,
+					nr_shrink - node_cnt - tree_cnt);
 			write_unlock(&et->lock);
 		}
-		f2fs_bug_on(sbi, atomic_read(&et->node_cnt));
+
+		if (atomic_read(&et->node_cnt))
+			goto unlock_out;
+
 		list_del_init(&et->list);
 		radix_tree_delete(&eti->extent_tree_root, et->ino);
 		kmem_cache_free(extent_tree_slab, et);
···
 	return __shrink_extent_tree(sbi, nr_shrink, EX_BLOCK_AGE);
 }
 
-static unsigned int __destroy_extent_node(struct inode *inode,
-					enum extent_type type)
-{
-	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct extent_tree *et = F2FS_I(inode)->extent_tree[type];
-	unsigned int node_cnt = 0;
-
-	if (!et || !atomic_read(&et->node_cnt))
-		return 0;
-
-	write_lock(&et->lock);
-	node_cnt = __free_extent_tree(sbi, et);
-	write_unlock(&et->lock);
-
-	return node_cnt;
-}
-
 void f2fs_destroy_extent_node(struct inode *inode)
 {
 	__destroy_extent_node(inode, EX_READ);
···
 
 static void __drop_extent_tree(struct inode *inode, enum extent_type type)
 {
-	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct extent_tree *et = F2FS_I(inode)->extent_tree[type];
 	bool updated = false;
 
···
 		return;
 
 	write_lock(&et->lock);
-	__free_extent_tree(sbi, et);
 	if (type == EX_READ) {
 		set_inode_flag(inode, FI_NO_EXTENT);
 		if (et->largest.len) {
···
 		}
 	}
 	write_unlock(&et->lock);
+
+	__destroy_extent_node(inode, type);
+
 	if (updated)
 		f2fs_mark_inode_dirty_sync(inode, true);
 }
···
 	sbi->hot_data_age_threshold = DEF_HOT_DATA_AGE_THRESHOLD;
 	sbi->warm_data_age_threshold = DEF_WARM_DATA_AGE_THRESHOLD;
 	sbi->last_age_weight = LAST_AGE_WEIGHT;
+	sbi->max_read_extent_count = DEF_MAX_READ_EXTENT_COUNT;
 }
 
 int __init f2fs_create_extent_cache(void)

fs/f2fs/f2fs.h

···
 #define F2FS_FEATURE_CASEFOLD			0x00001000
 #define F2FS_FEATURE_COMPRESSION		0x00002000
 #define F2FS_FEATURE_RO				0x00004000
+#define F2FS_FEATURE_DEVICE_ALIAS		0x00008000
 
 #define __F2FS_HAS_FEATURE(raw_super, mask)				\
 	((raw_super->feature & cpu_to_le32(mask)) != 0)
···
 #define DEF_HOT_DATA_AGE_THRESHOLD	262144
 #define DEF_WARM_DATA_AGE_THRESHOLD	2621440
 
+/* default max read extent count per inode */
+#define DEF_MAX_READ_EXTENT_COUNT	10240
+
 /* extent cache type */
 enum extent_type {
 	EX_READ,
···
 #define NR_CURSEG_PERSIST_TYPE	(NR_CURSEG_DATA_TYPE + NR_CURSEG_NODE_TYPE)
 #define NR_CURSEG_TYPE		(NR_CURSEG_INMEM_TYPE + NR_CURSEG_PERSIST_TYPE)
 
-enum {
+enum log_type {
 	CURSEG_HOT_DATA	= 0,	/* directory entry blocks */
 	CURSEG_WARM_DATA,	/* data blocks */
 	CURSEG_COLD_DATA,	/* multimedia or GCed data blocks */
···
 	unsigned int segment_count;	/* total # of segments */
 	unsigned int main_segments;	/* # of segments in main area */
 	unsigned int reserved_segments;	/* # of reserved segments */
-	unsigned int additional_reserved_segments;/* reserved segs for IO align feature */
 	unsigned int ovp_segments;	/* # of overprovision segments */
 
 	/* a threshold to reclaim prefree segments */
···
 	/* for extent tree cache */
 	struct extent_tree_info extent_tree[NR_EXTENT_CACHES];
 	atomic64_t allocated_data_blocks;	/* for block age extent_cache */
+	unsigned int max_read_extent_count;	/* max read extent count per inode */
 
 	/* The threshold used for hot and warm data seperation*/
 	unsigned int hot_data_age_threshold;
···
 	unsigned int dirty_device;		/* for checkpoint data flush */
 	spinlock_t dev_lock;			/* protect dirty_device */
 	bool aligned_blksize;			/* all devices has the same logical blksize */
+	unsigned int first_zoned_segno;		/* first zoned segno */
 
 	/* For write statistics */
 	u64 sectors_written_start;
···
 #define F2FS_DIRSYNC_FL			0x00010000 /* dirsync behaviour (directories only) */
 #define F2FS_PROJINHERIT_FL		0x20000000 /* Create with parents projid */
 #define F2FS_CASEFOLD_FL		0x40000000 /* Casefolded file */
+#define F2FS_DEVICE_ALIAS_FL		0x80000000 /* File for aliasing a device */
 
 #define F2FS_QUOTA_DEFAULT_FL		(F2FS_NOATIME_FL | F2FS_IMMUTABLE_FL)
 
···
 
 /* Flags that are appropriate for non-directories/regular files. */
 #define F2FS_OTHER_FLMASK	(F2FS_NODUMP_FL | F2FS_NOATIME_FL)
+
+#define IS_DEVICE_ALIASING(inode)	(F2FS_I(inode)->i_flags & F2FS_DEVICE_ALIAS_FL)
 
 static inline __u32 f2fs_mask_flags(umode_t mode, __u32 flags)
 {
···
 loff_t max_file_blocks(struct inode *inode);
 void f2fs_quota_off_umount(struct super_block *sb);
 void f2fs_save_errors(struct f2fs_sb_info *sbi, unsigned char flag);
-void f2fs_handle_critical_error(struct f2fs_sb_info *sbi, unsigned char reason,
-							bool irq_context);
+void f2fs_handle_critical_error(struct f2fs_sb_info *sbi, unsigned char reason);
 void f2fs_handle_error(struct f2fs_sb_info *sbi, unsigned char error);
 void f2fs_handle_error_async(struct f2fs_sb_info *sbi, unsigned char error);
 int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover);
···
 			block_t old_addr, block_t new_addr,
 			unsigned char version, bool recover_curseg,
 			bool recover_newaddr);
-int f2fs_get_segment_temp(int seg_type);
+enum temp_type f2fs_get_segment_temp(struct f2fs_sb_info *sbi,
+						enum log_type seg_type);
 int f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 			block_t old_blkaddr, block_t *new_blkaddr,
 			struct f2fs_summary *sum, int type,
···
 int f2fs_lookup_journal_in_cursum(struct f2fs_journal *journal, int type,
 			unsigned int val, int alloc);
 void f2fs_flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc);
-int f2fs_fix_curseg_write_pointer(struct f2fs_sb_info *sbi);
-int f2fs_check_write_pointer(struct f2fs_sb_info *sbi);
+int f2fs_check_and_fix_write_pointer(struct f2fs_sb_info *sbi);
 int f2fs_build_segment_manager(struct f2fs_sb_info *sbi);
 void f2fs_destroy_segment_manager(struct f2fs_sb_info *sbi);
 int __init f2fs_create_segment_manager_caches(void);
···
 			enum page_type type, enum temp_type temp);
 unsigned int f2fs_usable_segs_in_sec(struct f2fs_sb_info *sbi);
 unsigned int f2fs_usable_blks_in_seg(struct f2fs_sb_info *sbi,
+			unsigned int segno);
+unsigned long long f2fs_get_section_mtime(struct f2fs_sb_info *sbi,
 			unsigned int segno);
 
 #define DEF_FRAGMENT_SIZE	4
···
  * debug.c
  */
 #ifdef CONFIG_F2FS_STAT_FS
+enum {
+	DEVSTAT_INUSE,
+	DEVSTAT_DIRTY,
+	DEVSTAT_FULL,
+	DEVSTAT_FREE,
+	DEVSTAT_PREFREE,
+	DEVSTAT_MAX,
+};
+
+struct f2fs_dev_stats {
+	unsigned int devstats[2][DEVSTAT_MAX];		/* 0: segs, 1: secs */
+};
+
 struct f2fs_stat_info {
 	struct list_head stat_list;
 	struct f2fs_sb_info *sbi;
···
 	unsigned int block_count[2];
 	unsigned int inplace_count;
 	unsigned long long base_mem, cache_mem, page_mem;
+	struct f2fs_dev_stats *dev_stats;
 };
 
 static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
···
 F2FS_FEATURE_FUNCS(casefold, CASEFOLD);
 F2FS_FEATURE_FUNCS(compression, COMPRESSION);
 F2FS_FEATURE_FUNCS(readonly, RO);
+F2FS_FEATURE_FUNCS(device_alias, DEVICE_ALIAS);
 
 #ifdef CONFIG_BLK_DEV_ZONED
 static inline bool f2fs_blkz_is_seq(struct f2fs_sb_info *sbi, int devi,

fs/f2fs/file.c

···
 
 	trace_f2fs_truncate_blocks_enter(inode, from);
 
+	if (IS_DEVICE_ALIASING(inode) && from) {
+		err = -EINVAL;
+		goto out_err;
+	}
+
 	free_from = (pgoff_t)F2FS_BLK_ALIGN(from);
 
 	if (free_from >= max_file_blocks(inode))
···
 	ipage = f2fs_get_node_page(sbi, inode->i_ino);
 	if (IS_ERR(ipage)) {
 		err = PTR_ERR(ipage);
+		goto out;
+	}
+
+	if (IS_DEVICE_ALIASING(inode)) {
+		struct extent_tree *et = F2FS_I(inode)->extent_tree[EX_READ];
+		struct extent_info ei = et->largest;
+		unsigned int i;
+
+		for (i = 0; i < ei.len; i++)
+			f2fs_invalidate_blocks(sbi, ei.blk + i);
+
+		dec_valid_block_count(sbi, inode, ei.len);
+		f2fs_update_time(sbi, REQ_TIME);
+
+		f2fs_put_page(ipage, 1);
 		goto out;
 	}
 
···
 	/* lastly zero out the first data page */
 	if (!err)
 		err = truncate_partial_data_page(inode, from, truncate_page);
-
+out_err:
 	trace_f2fs_truncate_blocks_exit(inode, err);
 	return err;
 }
···
 		return true;
 	if (f2fs_compressed_file(inode))
 		return true;
-	if (f2fs_has_inline_data(inode))
+	/*
+	 * only force direct read to use buffered IO, for direct write,
+	 * it expects inline data conversion before committing IO.
+	 */
+	if (f2fs_has_inline_data(inode) && rw == READ)
 		return true;
 
 	/* disallow direct IO if any of devices has unaligned blksize */
···
 		return -EPERM;
 
 	if ((attr->ia_valid & ATTR_SIZE)) {
-		if (!f2fs_is_compress_backend_ready(inode))
+		if (!f2fs_is_compress_backend_ready(inode) ||
+				IS_DEVICE_ALIASING(inode))
 			return -EOPNOTSUPP;
 		if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED) &&
 			!IS_ALIGNED(attr->ia_size,
···
 
 		map.m_len = sec_blks;
 next_alloc:
-		if (has_not_enough_free_secs(sbi, 0,
+		if (has_not_enough_free_secs(sbi, 0, f2fs_sb_has_blkzoned(sbi) ?
+			ZONED_PIN_SEC_REQUIRED_COUNT :
 			GET_SEC_FROM_SEG(sbi, overprovision_segments(sbi)))) {
 			f2fs_down_write(&sbi->gc_lock);
 			stat_inc_gc_call_count(sbi, FOREGROUND);
···
 		return -EIO;
 	if (!f2fs_is_checkpoint_ready(F2FS_I_SB(inode)))
 		return -ENOSPC;
-	if (!f2fs_is_compress_backend_ready(inode))
+	if (!f2fs_is_compress_backend_ready(inode) || IS_DEVICE_ALIASING(inode))
 		return -EOPNOTSUPP;
 
 	/* f2fs only support ->fallocate for regular file */
···
 	if (readonly)
 		goto out;
 
-	/* grab sb->s_umount to avoid racing w/ remount() */
+	/*
+	 * grab sb->s_umount to avoid racing w/ remount() and other shutdown
+	 * paths.
+	 */
 	if (need_lock)
-		down_read(&sbi->sb->s_umount);
+		down_write(&sbi->sb->s_umount);
 
 	f2fs_stop_gc_thread(sbi);
 	f2fs_stop_discard_thread(sbi);
···
 	clear_opt(sbi, DISCARD);
 
 	if (need_lock)
-		up_read(&sbi->sb->s_umount);
+		up_write(&sbi->sb->s_umount);
 
 	f2fs_update_time(sbi, REQ_TIME);
 out:
···
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	if (!S_ISREG(inode->i_mode) || f2fs_is_atomic_file(inode))
+	if (!S_ISREG(inode->i_mode))
 		return -EINVAL;
 
 	if (f2fs_readonly(sbi->sb))
···
 	struct f2fs_inode_info *fi = F2FS_I(inode);
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 
+	if (IS_DEVICE_ALIASING(inode))
+		return -EINVAL;
+
 	if (fi->i_gc_failures >= sbi->gc_pin_file_threshold) {
 		f2fs_warn(sbi, "%s: Enable GC = ino %lx after %x GC trials",
 			  __func__, inode->i_ino, fi->i_gc_failures);
···
 
 	if (f2fs_readonly(sbi->sb))
 		return -EROFS;
+
+	if (!pin && IS_DEVICE_ALIASING(inode))
+		return -EOPNOTSUPP;
 
 	ret = mnt_want_write_file(filp);
 	if (ret)
···
 	if (is_inode_flag_set(inode, FI_PIN_FILE))
 		pin = F2FS_I(inode)->i_gc_failures;
 	return put_user(pin, (u32 __user *)arg);
+}
+
+static int f2fs_ioc_get_dev_alias_file(struct file *filp, unsigned long arg)
+{
+	return put_user(IS_DEVICE_ALIASING(file_inode(filp)) ? 1 : 0,
+			(u32 __user *)arg);
 }
 
 int f2fs_precache_extents(struct inode *inode)
···
 		to_reserved = cluster_size - compr_blocks - reserved;
 
 		/* for the case all blocks in cluster were reserved */
-		if (to_reserved == 1) {
+		if (reserved && to_reserved == 1) {
 			dn->ofs_in_node += cluster_size;
 			goto next;
 		}
···
 		return f2fs_ioc_decompress_file(filp);
 	case F2FS_IOC_COMPRESS_FILE:
 		return f2fs_ioc_compress_file(filp);
+	case F2FS_IOC_GET_DEV_ALIAS_FILE:
+		return f2fs_ioc_get_dev_alias_file(filp, arg);
 	default:
 		return -ENOTTY;
 	}
···
 	else
 		return 0;
 
-	map.m_may_create = true;
+	if (!IS_DEVICE_ALIASING(inode))
+		map.m_may_create = true;
 	if (dio) {
 		map.m_seg_type = f2fs_rw_hint_to_seg_type(sbi,
 						inode->i_write_hint);
···
 {
 	struct inode *inode = iter->inode;
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	int seg_type = f2fs_rw_hint_to_seg_type(sbi, inode->i_write_hint);
-	enum temp_type temp = f2fs_get_segment_temp(seg_type);
+	enum log_type type = f2fs_rw_hint_to_seg_type(sbi, inode->i_write_hint);
+	enum temp_type temp = f2fs_get_segment_temp(sbi, type);
 
 	bio->bi_write_hint = f2fs_io_type_to_rw_hint(sbi, DATA, temp);
 	submit_bio(bio);
···
 	case F2FS_IOC_SET_COMPRESS_OPTION:
 	case F2FS_IOC_DECOMPRESS_FILE:
 	case F2FS_IOC_COMPRESS_FILE:
+	case F2FS_IOC_GET_DEV_ALIAS_FILE:
 		break;
 	default:
 		return -ENOIOCTLCMD;

fs/f2fs/gc.c

···
 
 	switch (sbi->gc_mode) {
 	case GC_IDLE_CB:
+	case GC_URGENT_LOW:
+	case GC_URGENT_MID:
 		gc_mode = GC_CB;
 		break;
 	case GC_IDLE_GREEDY:
···
 static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
-	unsigned int start = GET_SEG_FROM_SEC(sbi, secno);
 	unsigned long long mtime = 0;
 	unsigned int vblocks;
 	unsigned char age = 0;
 	unsigned char u;
-	unsigned int i;
 	unsigned int usable_segs_per_sec = f2fs_usable_segs_in_sec(sbi);
 
-	for (i = 0; i < usable_segs_per_sec; i++)
-		mtime += get_seg_entry(sbi, start + i)->mtime;
+	mtime = f2fs_get_section_mtime(sbi, segno);
+	f2fs_bug_on(sbi, mtime == INVALID_MTIME);
 	vblocks = get_valid_blocks(sbi, segno, true);
-
-	mtime = div_u64(mtime, usable_segs_per_sec);
 	vblocks = div_u64(vblocks, usable_segs_per_sec);
 
 	u = BLKS_TO_SEGS(sbi, vblocks * 100);
···
 				struct victim_sel_policy *p, unsigned int segno)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
-	unsigned int start = GET_SEG_FROM_SEC(sbi, secno);
 	unsigned long long mtime = 0;
-	unsigned int i;
 
 	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) {
 		if (p->gc_mode == GC_AT &&
···
 			return;
 	}
 
-	for (i = 0; i < SEGS_PER_SEC(sbi); i++)
-		mtime += get_seg_entry(sbi, start + i)->mtime;
-	mtime = div_u64(mtime, SEGS_PER_SEC(sbi));
+	mtime = f2fs_get_section_mtime(sbi, segno);
+	f2fs_bug_on(sbi, mtime == INVALID_MTIME);
 
 	/* Handle if the system time has changed by the user */
 	if (mtime < sit_i->min_mtime)

fs/f2fs/gc.h

···
 #define LIMIT_BOOST_ZONED_GC	25 /* percentage over total user space of boosted gc for zoned devices */
 #define DEF_MIGRATION_WINDOW_GRANULARITY_ZONED	3
 #define BOOST_GC_MULTIPLE	5
+#define ZONED_PIN_SEC_REQUIRED_COUNT	1
 
 #define DEF_GC_FAILED_PINNED_FILES	2048
 #define MAX_GC_FAILED_PINNED_FILES	USHRT_MAX

fs/f2fs/inode.c

···
 		return false;
 	}
 
+	if (IS_DEVICE_ALIASING(inode)) {
+		if (!f2fs_sb_has_device_alias(sbi)) {
+			f2fs_warn(sbi, "%s: inode (ino=%lx) has device alias flag, but the feature is off",
+				  __func__, inode->i_ino);
+			return false;
+		}
+		if (!f2fs_is_pinned_file(inode)) {
+			f2fs_warn(sbi, "%s: inode (ino=%lx) has device alias flag, but is not pinned",
+				  __func__, inode->i_ino);
+			return false;
+		}
+	}
+
 	return true;
 }
 
···
 		!is_inode_flag_set(inode, FI_DIRTY_INODE))
 		return 0;
 
-	if (!f2fs_is_checkpoint_ready(sbi))
+	if (!f2fs_is_checkpoint_ready(sbi)) {
+		f2fs_mark_inode_dirty_sync(inode, true);
 		return -ENOSPC;
+	}
 
 	/*
 	 * We need to balance fs here to prevent from producing dirty node pages
···
 	f2fs_bug_on(sbi, get_dirty_pages(inode));
 	f2fs_remove_dirty_inode(inode);
 
-	f2fs_destroy_extent_tree(inode);
+	if (!IS_DEVICE_ALIASING(inode))
+		f2fs_destroy_extent_tree(inode);
 
 	if (inode->i_nlink || is_bad_inode(inode))
 		goto no_delete;
···
 		err = 0;
 		goto retry;
 	}
+
+	if (IS_DEVICE_ALIASING(inode))
+		f2fs_destroy_extent_tree(inode);
 
 	if (err) {
 		f2fs_update_inode_page(inode);

fs/f2fs/node.c

···
 	if (err)
 		return err;
 
+	if (ni.blk_addr != NEW_ADDR &&
+		!f2fs_is_valid_blkaddr(sbi, ni.blk_addr, DATA_GENERIC_ENHANCE)) {
+		f2fs_err_ratelimited(sbi,
+			"nat entry is corrupted, run fsck to fix it, ino:%u, "
+			"nid:%u, blkaddr:%u", ni.ino, ni.nid, ni.blk_addr);
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
+		f2fs_handle_error(sbi, ERROR_INCONSISTENT_NAT);
+		return -EFSCORRUPTED;
+	}
+
 	/* Deallocate node address */
 	f2fs_invalidate_blocks(sbi, ni.blk_addr);
 	dec_valid_node_count(sbi, dn->inode, dn->nid == dn->inode->i_ino);
···
 	int i;
 	int idx = depth - 2;
 
-	nid[0] = le32_to_cpu(ri->i_nid[offset[0] - NODE_DIR1_BLOCK]);
+	nid[0] = get_nid(dn->inode_page, offset[0], true);
 	if (!nid[0])
 		return 0;
 
···
 
 skip_partial:
 	while (cont) {
-		dn.nid = le32_to_cpu(ri->i_nid[offset[0] - NODE_DIR1_BLOCK]);
+		dn.nid = get_nid(page, offset[0], true);
 		switch (offset[0]) {
 		case NODE_DIR1_BLOCK:
 		case NODE_DIR2_BLOCK:
···
 		}
 		if (err < 0)
 			goto fail;
-		if (offset[1] == 0 &&
-				ri->i_nid[offset[0] - NODE_DIR1_BLOCK]) {
+		if (offset[1] == 0 && get_nid(page, offset[0], true)) {
 			lock_page(page);
 			BUG_ON(page->mapping != NODE_MAPPING(sbi));
-			f2fs_wait_on_page_writeback(page, NODE, true, true);
-			ri->i_nid[offset[0] - NODE_DIR1_BLOCK] = 0;
-			set_page_dirty(page);
+			set_nid(page, offset[0], 0, true);
 			unlock_page(page);
 		}
 		offset[1] = 0;
···
 		err = -EFSCORRUPTED;
 		dec_valid_node_count(sbi, dn->inode, !ofs);
 		set_sbi_flag(sbi, SBI_NEED_FSCK);
-		f2fs_handle_error(sbi, ERROR_INVALID_BLKADDR);
+		f2fs_warn_ratelimited(sbi,
+			"f2fs_new_node_page: inconsistent nat entry, "
+			"ino:%u, nid:%u, blkaddr:%u, ver:%u, flag:%u",
+			new_ni.ino, new_ni.nid, new_ni.blk_addr,
+			new_ni.version, new_ni.flag);
+		f2fs_handle_error(sbi, ERROR_INCONSISTENT_NAT);
 		goto fail;
 	}
 #endif

fs/f2fs/recovery.c

···
 	 * and the f2fs is not read only, check and fix zoned block devices'
 	 * write pointer consistency.
 	 */
-	if (f2fs_sb_has_blkzoned(sbi) && !f2fs_readonly(sbi->sb)) {
-		int err2 = f2fs_fix_curseg_write_pointer(sbi);
-
-		if (!err2)
-			err2 = f2fs_check_write_pointer(sbi);
-		if (err2)
-			err = err2;
+	if (!err) {
+		err = f2fs_check_and_fix_write_pointer(sbi);
 		ret = err;
 	}
 

fs/f2fs/segment.c

···
 						wait_list, issued);
 			return 0;
 		}
-
-		/*
-		 * Issue discard for conventional zones only if the device
-		 * supports discard.
-		 */
-		if (!bdev_max_discard_sectors(bdev))
-			return -EOPNOTSUPP;
 	}
 #endif
+
+	/*
+	 * stop issuing discard for any of below cases:
+	 * 1. device is conventional zone, but it doesn't support discard.
+	 * 2. device is regulare device, after snapshot it doesn't support
+	 * discard.
+	 */
+	if (!bdev_max_discard_sectors(bdev))
+		return -EOPNOTSUPP;
 
 	trace_f2fs_issue_discard(bdev, dc->di.start, dc->di.len);
 
···
 		if (sbi->blkzone_alloc_policy == BLKZONE_ALLOC_PRIOR_CONV || pinning)
 			segno = 0;
 		else
-			segno = max(first_zoned_segno(sbi), *newseg);
+			segno = max(sbi->first_zoned_segno, *newseg);
 		hint = GET_SEC_FROM_SEG(sbi, segno);
 	}
 #endif
···
 	if (secno >= MAIN_SECS(sbi) && f2fs_sb_has_blkzoned(sbi)) {
 		/* Write only to sequential zones */
 		if (sbi->blkzone_alloc_policy == BLKZONE_ALLOC_ONLY_SEQ) {
-			hint = GET_SEC_FROM_SEG(sbi, first_zoned_segno(sbi));
+			hint = GET_SEC_FROM_SEG(sbi, sbi->first_zoned_segno);
 			secno = find_next_zero_bit(free_i->free_secmap, MAIN_SECS(sbi), hint);
 		} else
 			secno = find_first_zero_bit(free_i->free_secmap,
···
 	struct f2fs_summary_block *sum_node;
 	struct page *sum_page;
 
-	write_sum_page(sbi, curseg->sum_blk, GET_SUM_BLOCK(sbi, curseg->segno));
+	if (curseg->inited)
+		write_sum_page(sbi, curseg->sum_blk, GET_SUM_BLOCK(sbi, curseg->segno));
 
 	__set_test_and_inuse(sbi, new_segno);
 
···
 
 	if (f2fs_sb_has_blkzoned(sbi) && err == -EAGAIN && gc_required) {
 		f2fs_down_write(&sbi->gc_lock);
-		err = f2fs_gc_range(sbi, 0, GET_SEGNO(sbi, FDEV(0).end_blk), true, 1);
+		err = f2fs_gc_range(sbi, 0, GET_SEGNO(sbi, FDEV(0).end_blk),
+				true, ZONED_PIN_SEC_REQUIRED_COUNT);
 		f2fs_up_write(&sbi->gc_lock);
 
 		gc_required = false;
···
 	}
 }
 
-int f2fs_get_segment_temp(int seg_type)
+enum temp_type f2fs_get_segment_temp(struct f2fs_sb_info *sbi,
+						enum log_type type)
 {
-	if (IS_HOT(seg_type))
-		return HOT;
-	else if (IS_WARM(seg_type))
-		return WARM;
-	return COLD;
+	struct curseg_info *curseg = CURSEG_I(sbi, type);
+	enum temp_type temp = COLD;
+
+	switch (curseg->seg_type) {
+	case CURSEG_HOT_NODE:
+	case CURSEG_HOT_DATA:
+		temp = HOT;
+		break;
+	case CURSEG_WARM_NODE:
+	case CURSEG_WARM_DATA:
+		temp = WARM;
+		break;
+	case CURSEG_COLD_NODE:
+	case CURSEG_COLD_DATA:
+		temp = COLD;
+		break;
+	default:
+		f2fs_bug_on(sbi, 1);
+	}
+
+	return temp;
 }
 
 static int __get_segment_type(struct f2fs_io_info *fio)
 {
-	int type = 0;
+	enum log_type type = CURSEG_HOT_DATA;
 
 	switch (F2FS_OPTION(fio->sbi).active_logs) {
 	case 2:
···
 		f2fs_bug_on(fio->sbi, true);
 	}
 
-	fio->temp = f2fs_get_segment_temp(type);
+	fio->temp = f2fs_get_segment_temp(fio->sbi, type);
 
 	return type;
 }
···
 	}
 }
 
+static int log_type_to_seg_type(enum log_type type)
+{
+	int seg_type = CURSEG_COLD_DATA;
+
+	switch (type) {
+	case CURSEG_HOT_DATA:
+	case CURSEG_WARM_DATA:
+	case CURSEG_COLD_DATA:
+	case CURSEG_HOT_NODE:
+	case CURSEG_WARM_NODE:
+	case CURSEG_COLD_NODE:
+		seg_type = (int)type;
+		break;
+	case CURSEG_COLD_DATA_PINNED:
+	case CURSEG_ALL_DATA_ATGC:
+		seg_type = CURSEG_COLD_DATA;
+		break;
+	default:
+		break;
+	}
+	return seg_type;
+}
+
 static void do_write_page(struct f2fs_summary *sum, struct f2fs_io_info *fio)
 {
-	int type = __get_segment_type(fio);
-	bool keep_order = (f2fs_lfs_mode(fio->sbi) && type == CURSEG_COLD_DATA);
+	enum log_type type = __get_segment_type(fio);
+	int seg_type = log_type_to_seg_type(type);
+	bool keep_order = (f2fs_lfs_mode(fio->sbi) &&
+				seg_type == CURSEG_COLD_DATA);
 
 	if (keep_order)
 		f2fs_down_read(&fio->sbi->io_order_lock);
···
 		}
 	}
 
-	f2fs_bug_on(sbi, !IS_DATASEG(type));
 	curseg = CURSEG_I(sbi, type);
+	f2fs_bug_on(sbi, !IS_DATASEG(curseg->seg_type));
 
 	mutex_lock(&curseg->curseg_mutex);
 	down_write(&sit_i->sentry_lock);
···
 				sizeof(struct f2fs_journal), GFP_KERNEL);
 		if (!array[i].journal)
 			return -ENOMEM;
-		if (i < NR_PERSISTENT_LOG)
-			array[i].seg_type = CURSEG_HOT_DATA + i;
-		else if (i == CURSEG_COLD_DATA_PINNED)
-			array[i].seg_type = CURSEG_COLD_DATA;
-		else if (i == CURSEG_ALL_DATA_ATGC)
-			array[i].seg_type = CURSEG_COLD_DATA;
+		array[i].seg_type = log_type_to_seg_type(i);
 		reset_curseg_fields(&array[i]);
 	}
 	return restore_curseg_summaries(sbi);
···
 	return 0;
 }
 
-static int fix_curseg_write_pointer(struct f2fs_sb_info *sbi, int type)
+static int do_fix_curseg_write_pointer(struct f2fs_sb_info *sbi, int type)
 {
 	struct curseg_info *cs = CURSEG_I(sbi, type);
 	struct f2fs_dev_info *zbd;
···
 	return 0;
 }
 
-int f2fs_fix_curseg_write_pointer(struct f2fs_sb_info *sbi)
+static int fix_curseg_write_pointer(struct f2fs_sb_info *sbi)
 {
 	int i, ret;
 
 	for (i = 0; i < NR_PERSISTENT_LOG; i++) {
-		ret = fix_curseg_write_pointer(sbi, i);
+		ret = do_fix_curseg_write_pointer(sbi, i);
 		if (ret)
 			return ret;
 	}
···
 	return check_zone_write_pointer(args->sbi, args->fdev, zone);
 }
 
-int f2fs_check_write_pointer(struct f2fs_sb_info *sbi)
+static int check_write_pointer(struct f2fs_sb_info *sbi)
 {
 	int i, ret;
 	struct check_zone_write_pointer_args args;
···
 	}
 
 	return 0;
+}
+
+int f2fs_check_and_fix_write_pointer(struct f2fs_sb_info *sbi)
+{
+	int ret;
+
+	if (!f2fs_sb_has_blkzoned(sbi) || f2fs_readonly(sbi->sb))
+		return 0;
+
+	f2fs_notice(sbi, "Checking entire write pointers");
+	ret = fix_curseg_write_pointer(sbi);
+	if (!ret)
+		ret = check_write_pointer(sbi);
+	return ret;
 }
 
 /*
···
 	return BLKS_PER_SEG(sbi);
 }
 #else
-int f2fs_fix_curseg_write_pointer(struct f2fs_sb_info *sbi)
-{
-	return 0;
-}
-
-int f2fs_check_write_pointer(struct f2fs_sb_info *sbi)
+int f2fs_check_and_fix_write_pointer(struct f2fs_sb_info *sbi)
 {
 	return 0;
 }
···
 	return SEGS_PER_SEC(sbi);
 }
 
+unsigned long long f2fs_get_section_mtime(struct f2fs_sb_info *sbi,
+	unsigned int segno)
+{
+	unsigned int usable_segs_per_sec = f2fs_usable_segs_in_sec(sbi);
+	unsigned int secno = 0, start = 0;
+	unsigned int total_valid_blocks = 0;
+	unsigned long long mtime = 0;
+	unsigned int i = 0;
+
+	secno = GET_SEC_FROM_SEG(sbi, segno);
+	start = GET_SEG_FROM_SEC(sbi, secno);
+
+	if (!__is_large_section(sbi))
+		return get_seg_entry(sbi, start + i)->mtime;
+
+	for (i = 0; i < usable_segs_per_sec; i++) {
+		/* for large section, only check the mtime of valid segments */
+		struct seg_entry *se = get_seg_entry(sbi, start+i);
+
+		mtime += se->mtime * se->valid_blocks;
+		total_valid_blocks += se->valid_blocks;
+	}
+
+	if (total_valid_blocks == 0)
+		return INVALID_MTIME;
+
+	return div_u64(mtime, total_valid_blocks);
+}
+
 /*
  * Update min, max modified time for cost-benefit GC algorithm
  */
···
 	sit_i->min_mtime = ULLONG_MAX;
 
 	for (segno = 0; segno < MAIN_SEGS(sbi); segno += SEGS_PER_SEC(sbi)) {
-		unsigned int i;
 		unsigned long long mtime = 0;
 
-		for (i = 0; i < SEGS_PER_SEC(sbi); i++)
-			mtime += get_seg_entry(sbi, segno + i)->mtime;
-
-		mtime = div_u64(mtime, SEGS_PER_SEC(sbi));
+		mtime = f2fs_get_section_mtime(sbi, segno);
 
 		if (sit_i->min_mtime > mtime)
 			sit_i->min_mtime = mtime;

fs/f2fs/segment.h

···
 #define F2FS_MIN_SEGMENTS	9 /* SB + 2 (CP + SIT + NAT) + SSA + MAIN */
 #define F2FS_MIN_META_SEGMENTS	8 /* SB + 2 (CP + SIT + NAT) + SSA */
 
+#define INVALID_MTIME ULLONG_MAX /* no valid blocks in a segment/section */
+
 /* L: Logical segment # in volume, R: Relative segment # in main area */
 #define GET_L2R_SEGNO(free_i, segno)	((segno) - (free_i)->start_segno)
 #define GET_R2L_SEGNO(free_i, segno)	((segno) + (free_i)->start_segno)
···
 {
 	f2fs_bug_on(sbi, seg_type >= NR_PERSISTENT_LOG);
 }
-
-#define IS_HOT(t)	((t) == CURSEG_HOT_NODE || (t) == CURSEG_HOT_DATA)
-#define IS_WARM(t)	((t) == CURSEG_WARM_NODE || (t) == CURSEG_WARM_DATA)
-#define IS_COLD(t)	((t) == CURSEG_COLD_NODE || (t) == CURSEG_COLD_DATA)
 
 #define IS_CURSEG(sbi, seg)						\
 	(((seg) == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno) ||	\
···
 
 static inline unsigned int reserved_segments(struct f2fs_sb_info *sbi)
 {
-	return SM_I(sbi)->reserved_segments +
-			SM_I(sbi)->additional_reserved_segments;
+	return SM_I(sbi)->reserved_segments;
 }
 
 static inline unsigned int free_sections(struct f2fs_sb_info *sbi)
···
 }
 
 static inline bool has_curseg_enough_space(struct f2fs_sb_info *sbi,
-			unsigned int node_blocks, unsigned int dent_blocks)
+			unsigned int node_blocks, unsigned int data_blocks,
+			unsigned int dent_blocks)
 {
 
-	unsigned segno, left_blocks;
+	unsigned int segno, left_blocks, blocks;
 	int i;
 
-	/* check current node sections in the worst case. */
-	for (i = CURSEG_HOT_NODE; i <= CURSEG_COLD_NODE; i++) {
+	/* check current data/node sections in the worst case. */
+	for (i = CURSEG_HOT_DATA; i < NR_PERSISTENT_LOG; i++) {
 		segno = CURSEG_I(sbi, i)->segno;
 		left_blocks = CAP_BLKS_PER_SEC(sbi) -
 				get_ckpt_valid_blocks(sbi, segno, true);
-		if (node_blocks > left_blocks)
+
+		blocks = i <= CURSEG_COLD_DATA ? data_blocks : node_blocks;
+		if (blocks > left_blocks)
 			return false;
 	}
 
···
 }
 
 /*
- * calculate needed sections for dirty node/dentry
- * and call has_curseg_enough_space
+ * calculate needed sections for dirty node/dentry and call
+ * has_curseg_enough_space, please note that, it needs to account
+ * dirty data as well in lfs mode when checkpoint is disabled.
  */
 static inline void __get_secs_required(struct f2fs_sb_info *sbi,
 		unsigned int *lower_p, unsigned int *upper_p, bool *curseg_p)
···
 					get_pages(sbi, F2FS_DIRTY_DENTS) +
 					get_pages(sbi, F2FS_DIRTY_IMETA);
 	unsigned int total_dent_blocks = get_pages(sbi, F2FS_DIRTY_DENTS);
+	unsigned int total_data_blocks = 0;
 	unsigned int node_secs = total_node_blocks / CAP_BLKS_PER_SEC(sbi);
 	unsigned int dent_secs = total_dent_blocks / CAP_BLKS_PER_SEC(sbi);
+	unsigned int data_secs = 0;
 	unsigned int node_blocks = total_node_blocks % CAP_BLKS_PER_SEC(sbi);
 	unsigned int dent_blocks = total_dent_blocks % CAP_BLKS_PER_SEC(sbi);
+	unsigned int data_blocks = 0;
+
+	if (f2fs_lfs_mode(sbi) &&
+		unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) {
+		total_data_blocks = get_pages(sbi, F2FS_DIRTY_DATA);
+		data_secs = total_data_blocks / CAP_BLKS_PER_SEC(sbi);
+		data_blocks = total_data_blocks % CAP_BLKS_PER_SEC(sbi);
+	}
 
 	if (lower_p)
-		*lower_p = node_secs + dent_secs;
+		*lower_p = node_secs + dent_secs + data_secs;
 	if (upper_p)
 		*upper_p = node_secs + dent_secs +
-			(node_blocks ? 1 : 0) + (dent_blocks ? 1 : 0);
+			(node_blocks ? 1 : 0) + (dent_blocks ? 1 : 0) +
+			(data_blocks ? 1 : 0);
 	if (curseg_p)
 		*curseg_p = has_curseg_enough_space(sbi,
-				node_blocks, dent_blocks);
+				node_blocks, data_blocks, dent_blocks);
 }
 
 static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi,
···
 	return !has_not_enough_free_secs(sbi, freed, needed);
 }
 
+static inline bool has_enough_free_blks(struct f2fs_sb_info *sbi)
+{
+	unsigned int total_free_blocks = 0;
+	unsigned int avail_user_block_count;
+
+	spin_lock(&sbi->stat_lock);
+
+	avail_user_block_count = get_available_block_count(sbi, NULL, true);
+	total_free_blocks = avail_user_block_count - (unsigned int)valid_user_blocks(sbi);
+
+	spin_unlock(&sbi->stat_lock);
+
+	return total_free_blocks > 0;
+}
+
 static inline bool f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
 {
 	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
 		return true;
 	if (likely(has_enough_free_secs(sbi, 0, 0)))
+		return true;
+	if (!f2fs_lfs_mode(sbi) &&
+		likely(has_enough_free_blks(sbi)))
 		return true;
 	return false;
 }
···
 wake_up:
 	dcc->discard_wake = true;
 	wake_up_interruptible_all(&dcc->discard_wait_queue);
-}
-
-static inline unsigned int first_zoned_segno(struct f2fs_sb_info *sbi)
-{
-	int devi;
-
-	for (devi = 0; devi < sbi->s_ndevs; devi++)
-		if (bdev_is_zoned(FDEV(devi).bdev))
-			return GET_SEGNO(sbi, FDEV(devi).start_blk);
-	return 0;
 }

fs/f2fs/super.c

···
 	Opt_mode,
 	Opt_fault_injection,
 	Opt_fault_type,
+	Opt_lazytime,
+	Opt_nolazytime,
 	Opt_quota,
 	Opt_noquota,
 	Opt_usrquota,
···
 	{Opt_mode, "mode=%s"},
 	{Opt_fault_injection, "fault_injection=%u"},
 	{Opt_fault_type, "fault_type=%u"},
+	{Opt_lazytime, "lazytime"},
+	{Opt_nolazytime, "nolazytime"},
 	{Opt_quota, "quota"},
 	{Opt_noquota, "noquota"},
 	{Opt_usrquota, "usrquota"},
···
 			set_opt(sbi, READ_EXTENT_CACHE);
 			break;
 		case Opt_noextent_cache:
+			if (F2FS_HAS_FEATURE(sbi, F2FS_FEATURE_DEVICE_ALIAS)) {
+				f2fs_err(sbi, "device aliasing requires extent cache");
+				return -EINVAL;
+			}
 			clear_opt(sbi, READ_EXTENT_CACHE);
 			break;
 		case Opt_noinline_data:
···
 			f2fs_info(sbi, "fault_type options not supported");
 			break;
 #endif
+		case Opt_lazytime:
+			sb->s_flags |= SB_LAZYTIME;
+			break;
+		case Opt_nolazytime:
+			sb->s_flags &= ~SB_LAZYTIME;
+			break;
 #ifdef CONFIG_QUOTA
 		case Opt_quota:
 		case Opt_usrquota:
···
 				break;
 			}
 
-			strcpy(ext[ext_cnt], name);
+			ret = strscpy(ext[ext_cnt], name);
+			if (ret < 0) {
+				kfree(name);
+				return ret;
+			}
 			F2FS_OPTION(sbi).compress_ext_cnt++;
 			kfree(name);
 			break;
···
 				break;
 			}
 
-			strcpy(noext[noext_cnt], name);
+			ret = strscpy(noext[noext_cnt], name);
+			if (ret < 0) {
+				kfree(name);
+				return ret;
+			}
 			F2FS_OPTION(sbi).nocompress_ext_cnt++;
 			kfree(name);
 			break;
···
 
 static int f2fs_unfreeze(struct super_block *sb)
 {
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+
+	/*
+	 * It will update discard_max_bytes of mounted lvm device to zero
+	 * after creating snapshot on this lvm device, let's drop all
+	 * remained discards.
+	 * We don't need to disable real-time discard because discard_max_bytes
+	 * will recover after removal of snapshot.
+	 */
+	if (test_opt(sbi, DISCARD) && !f2fs_hw_support_discard(sbi))
+		f2fs_issue_discard_timeout(sbi);
+
 	clear_sbi_flag(F2FS_SB(sb), SBI_IS_FREEZING);
 	return 0;
 }
···
 		}
 	}
 
+	adjust_unusable_cap_perc(sbi);
 	if (enable_checkpoint == !!test_opt(sbi, DISABLE_CHECKPOINT)) {
 		if (test_opt(sbi, DISABLE_CHECKPOINT)) {
 			err = f2fs_disable_checkpoint(sbi);
···
 		(test_opt(sbi, POSIX_ACL) ? SB_POSIXACL : 0);
 
 	limit_reserve_root(sbi);
-	adjust_unusable_cap_perc(sbi);
 	*flags = (*flags & ~SB_LAZYTIME) | (sb->s_flags & SB_LAZYTIME);
 	return 0;
 restore_checkpoint:
···
 	 * fit within U32_MAX + 1 data units.
 	 */
 
-	result = min(result, F2FS_BYTES_TO_BLK(((loff_t)U32_MAX + 1) * 4096));
+	result = umin(result, F2FS_BYTES_TO_BLK(((loff_t)U32_MAX + 1) * 4096));
 
 	return result;
 }
···
 		|| system_state == SYSTEM_RESTART;
 }
 
-void f2fs_handle_critical_error(struct f2fs_sb_info *sbi, unsigned char reason,
-							bool irq_context)
+void f2fs_handle_critical_error(struct f2fs_sb_info *sbi, unsigned char reason)
 {
 	struct super_block *sb = sbi->sb;
 	bool shutdown = reason == STOP_CP_REASON_SHUTDOWN;
···
 	if (!f2fs_hw_is_readonly(sbi)) {
 		save_stop_reason(sbi, reason);
 
-		if (irq_context && !shutdown)
-			schedule_work(&sbi->s_error_work);
-		else
-			f2fs_record_stop_reason(sbi);
+		/*
+		 * always create an asynchronous task to record stop_reason
+		 * in order to avoid potential deadlock when running into
+		 * f2fs_record_stop_reason() synchronously.
+		 */
+		schedule_work(&sbi->s_error_work);
 	}
 
 	/*
···
 					struct f2fs_sb_info, s_error_work);
 
 	f2fs_record_stop_reason(sbi);
+}
+
+static inline unsigned int get_first_zoned_segno(struct f2fs_sb_info *sbi)
+{
+	int devi;
+
+	for (devi = 0; devi < sbi->s_ndevs; devi++)
+		if (bdev_is_zoned(FDEV(devi).bdev))
+			return GET_SEGNO(sbi, FDEV(devi).start_blk);
+	return 0;
 }
 
 static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
···
 	/* For write statistics */
 	sbi->sectors_written_start = f2fs_get_sectors_written(sbi);
 
+	/* get segno of first zoned block device */
+	sbi->first_zoned_segno = get_first_zoned_segno(sbi);
+
 	/* Read accumulated write IO statistics if exists */
 	seg_i = CURSEG_I(sbi, CURSEG_HOT_NODE);
 	if (__exist_node_summaries(sbi))
···
 reset_checkpoint:
 	/*
 	 * If the f2fs is not readonly and fsync data recovery succeeds,
-	 * check zoned block devices' write pointer consistency.
+	 * write pointer consistency of cursegs and other zones are already
+	 * checked and fixed during recovery. However, if recovery fails,
+	 * write pointers are left untouched, and retry-mount should check
+	 * them here.
 	 */
-	if (f2fs_sb_has_blkzoned(sbi) && !f2fs_readonly(sb)) {
-		int err2;
-
-		f2fs_notice(sbi, "Checking entire write pointers");
-		err2 = f2fs_check_write_pointer(sbi);
-		if (err2)
-			err = err2;
-	}
+	if (skip_recovery)
+		err = f2fs_check_and_fix_write_pointer(sbi);
 	if (err)
 		goto free_meta;
+
+	/* f2fs_recover_fsync_data() cleared this already */
+	clear_sbi_flag(sbi, SBI_POR_DOING);
 
 	err = f2fs_init_inmem_curseg(sbi);
 	if (err)
 		goto sync_free_meta;
-
-	/* f2fs_recover_fsync_data() cleared this already */
-	clear_sbi_flag(sbi, SBI_POR_DOING);
 
 	if (test_opt(sbi, DISABLE_CHECKPOINT)) {
 		err = f2fs_disable_checkpoint(sbi);
···
 	err = f2fs_init_shrinker();
 	if (err)
 		goto free_sysfs;
-	err = register_filesystem(&f2fs_fs_type);
-	if (err)
-		goto free_shrinker;
 	f2fs_create_root_stats();
 	err = f2fs_init_post_read_processing();
 	if (err)
···
 	err = f2fs_create_casefold_cache();
 	if (err)
 		goto free_compress_cache;
+	err = register_filesystem(&f2fs_fs_type);
+	if (err)
+		goto free_casefold_cache;
 	return 0;
+free_casefold_cache:
+	f2fs_destroy_casefold_cache();
 free_compress_cache:
 	f2fs_destroy_compress_cache();
 free_compress_mempool:
···
 	f2fs_destroy_post_read_processing();
 free_root_stats:
 	f2fs_destroy_root_stats();
-	unregister_filesystem(&f2fs_fs_type);
-free_shrinker:
 	f2fs_exit_shrinker();
 free_sysfs:
 	f2fs_exit_sysfs();
···
 
 static void __exit exit_f2fs_fs(void)
 {
+	unregister_filesystem(&f2fs_fs_type);
 	f2fs_destroy_casefold_cache();
 	f2fs_destroy_compress_cache();
 	f2fs_destroy_compress_mempool();
···
 	f2fs_destroy_iostat_processing();
 	f2fs_destroy_post_read_processing();
 	f2fs_destroy_root_stats();
-	unregister_filesystem(&f2fs_fs_type);
 	f2fs_exit_shrinker();
 	f2fs_exit_sysfs();
 	f2fs_destroy_garbage_collection_cache();

fs/f2fs/sysfs.c

···
 	if (a->struct_type == RESERVED_BLOCKS) {
 		spin_lock(&sbi->stat_lock);
 		if (t > (unsigned long)(sbi->user_block_count -
-				F2FS_OPTION(sbi).root_reserved_blocks -
-				SEGS_TO_BLKS(sbi,
-				SM_I(sbi)->additional_reserved_segments))) {
+				F2FS_OPTION(sbi).root_reserved_blocks)) {
 			spin_unlock(&sbi->stat_lock);
 			return -EINVAL;
 		}
···
 		return count;
 	}
 
+	if (!strcmp(a->attr.name, "max_read_extent_count")) {
+		if (t > UINT_MAX)
+			return -EINVAL;
+		*ui = (unsigned int)t;
+		return count;
+	}
+
 	if (!strcmp(a->attr.name, "ipu_policy")) {
 		if (t >= BIT(F2FS_IPU_MAX))
 			return -EINVAL;
···
 F2FS_SBI_GENERAL_RW_ATTR(hot_data_age_threshold);
 F2FS_SBI_GENERAL_RW_ATTR(warm_data_age_threshold);
 F2FS_SBI_GENERAL_RW_ATTR(last_age_weight);
+/* read extent cache */
+F2FS_SBI_GENERAL_RW_ATTR(max_read_extent_count);
 #ifdef CONFIG_BLK_DEV_ZONED
 F2FS_SBI_GENERAL_RO_ATTR(unusable_blocks_per_sec);
 F2FS_SBI_GENERAL_RW_ATTR(blkzone_alloc_policy);
···
 	ATTR_LIST(hot_data_age_threshold),
 	ATTR_LIST(warm_data_age_threshold),
 	ATTR_LIST(last_age_weight),
+	ATTR_LIST(max_read_extent_count),
 	NULL,
 };
 ATTRIBUTE_GROUPS(f2fs);
···
 F2FS_SB_FEATURE_RO_ATTR(casefold, CASEFOLD);
 F2FS_SB_FEATURE_RO_ATTR(compression, COMPRESSION);
 F2FS_SB_FEATURE_RO_ATTR(readonly, RO);
+F2FS_SB_FEATURE_RO_ATTR(device_alias, DEVICE_ALIAS);
 
 static struct attribute *f2fs_sb_feat_attrs[] = {
 	ATTR_LIST(sb_encryption),
···
 	ATTR_LIST(sb_casefold),
 	ATTR_LIST(sb_compression),
 	ATTR_LIST(sb_readonly),
+	ATTR_LIST(sb_device_alias),
 	NULL,
 };
 ATTRIBUTE_GROUPS(f2fs_sb_feat);

include/linux/f2fs_fs.h

···
 #define NEW_ADDR		((block_t)-1)	/* used as block_t addresses */
 #define COMPRESS_ADDR		((block_t)-2)	/* used as compressed data flag */
 
-#define F2FS_BYTES_TO_BLK(bytes)	((bytes) >> F2FS_BLKSIZE_BITS)
-#define F2FS_BLK_TO_BYTES(blk)		((blk) << F2FS_BLKSIZE_BITS)
+#define F2FS_BLKSIZE_MASK		(F2FS_BLKSIZE - 1)
+#define F2FS_BYTES_TO_BLK(bytes)	((unsigned long long)(bytes) >> F2FS_BLKSIZE_BITS)
+#define F2FS_BLK_TO_BYTES(blk)		((unsigned long long)(blk) << F2FS_BLKSIZE_BITS)
 #define F2FS_BLK_END_BYTES(blk)		(F2FS_BLK_TO_BYTES(blk + 1) - 1)
-#define F2FS_BLK_ALIGN(x)			(F2FS_BYTES_TO_BLK((x) + F2FS_BLKSIZE - 1))
+#define F2FS_BLK_ALIGN(x)		(F2FS_BYTES_TO_BLK((x) + F2FS_BLKSIZE - 1))
 
 /* 0, 1(node nid), 2(meta nid) are reserved node id */
 #define F2FS_RESERVED_NODE_NUM		3

include/uapi/linux/f2fs.h

···
 #define F2FS_IOC_DECOMPRESS_FILE	_IO(F2FS_IOCTL_MAGIC, 23)
 #define F2FS_IOC_COMPRESS_FILE		_IO(F2FS_IOCTL_MAGIC, 24)
 #define F2FS_IOC_START_ATOMIC_REPLACE	_IO(F2FS_IOCTL_MAGIC, 25)
+#define F2FS_IOC_GET_DEV_ALIAS_FILE	_IOR(F2FS_IOCTL_MAGIC, 26, __u32)
 
 /*
  * should be same as XFS_IOC_GOINGDOWN.