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
  ...

1 files changed, 73 insertions, 28 deletions

diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 87ab5696bd48..fc7d463dee15 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -150,6 +150,8 @@ enum {
 	Opt_mode,
 	Opt_fault_injection,
 	Opt_fault_type,
+	Opt_lazytime,
+	Opt_nolazytime,
 	Opt_quota,
 	Opt_noquota,
 	Opt_usrquota,
@@ -226,6 +228,8 @@ static match_table_t f2fs_tokens = {
 	{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"},
@@ -834,6 +838,10 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount)
 			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:
@@ -918,6 +926,12 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount)
 			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:
@@ -1158,7 +1172,11 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount)
 				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;
@@ -1187,7 +1205,11 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount)
 				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;
@@ -1738,6 +1760,18 @@ static int f2fs_freeze(struct super_block *sb)
 
 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;
 }
@@ -2474,6 +2508,7 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 		}
 	}
 
+	adjust_unusable_cap_perc(sbi);
 	if (enable_checkpoint == !!test_opt(sbi, DISABLE_CHECKPOINT)) {
 		if (test_opt(sbi, DISABLE_CHECKPOINT)) {
 			err = f2fs_disable_checkpoint(sbi);
@@ -2518,7 +2553,6 @@ skip:
 		(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:
@@ -3322,7 +3356,7 @@ loff_t max_file_blocks(struct inode *inode)
 	 * 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;
 }
@@ -4155,8 +4189,7 @@ static bool system_going_down(void)
 		|| 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;
@@ -4168,10 +4201,12 @@ void f2fs_handle_critical_error(struct f2fs_sb_info *sbi, unsigned char reason,
 	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);
 	}
 
 	/*
@@ -4217,6 +4252,16 @@ static void f2fs_record_error_work(struct work_struct *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)
 {
 	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
@@ -4617,6 +4662,9 @@ try_onemore:
 	/* 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))
@@ -4738,26 +4786,23 @@ try_onemore:
 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);
 		if (err)
@@ -4991,9 +5036,6 @@ static int __init init_f2fs_fs(void)
 	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)
@@ -5016,7 +5058,12 @@ static int __init init_f2fs_fs(void)
 	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:
@@ -5031,8 +5078,6 @@ free_post_read:
 	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();
@@ -5056,6 +5101,7 @@ fail:
 
 static void __exit exit_f2fs_fs(void)
 {
+	unregister_filesystem(&f2fs_fs_type);
 	f2fs_destroy_casefold_cache();
 	f2fs_destroy_compress_cache();
 	f2fs_destroy_compress_mempool();
@@ -5064,7 +5110,6 @@ static void __exit exit_f2fs_fs(void)
 	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();