Skip to contentSkip to conversation

f2fs: checkpoint disabling

PATCH In review Jaegeuk Kim 1 participant 2 messages Sep 26, 2018 – Sep 28, 2018

Subsystems (2)
lore
Share & tools
lore

Jaegeuk Kim

author maintainer
#0 lore raw

PATCH 1/1 f2fs: checkpoint disabling

From: Daniel Rosenberg <[email protected]>

Note that, it requires "f2fs: return correct errno in f2fs_gc".

This adds a lightweight non-persistent snapshotting scheme to f2fs.

To use, mount with the option checkpoint=disable, and to return to normal operation, remount with checkpoint=enable. If the filesystem is shut down before remounting with checkpoint=enable, it will revert back to its apparent state when it was first mounted with checkpoint=disable. This is useful for situations where you wish to be able to roll back the state of the disk in case of some critical failure.

---
 Documentation/filesystems/f2fs.txt |   5 ++
 fs/f2fs/checkpoint.c               |  16 ++++
 fs/f2fs/data.c                     |  14 +++-
 fs/f2fs/debug.c                    |   3 +-
 fs/f2fs/f2fs.h                     |  15 +++-
 fs/f2fs/file.c                     |  12 +++
 fs/f2fs/gc.c                       |   9 ++-
 fs/f2fs/inode.c                    |   6 +-
 fs/f2fs/namei.c                    |  19 +++++
 fs/f2fs/segment.c                  |  75 +++++++++++++++++-
 fs/f2fs/segment.h                  |  15 ++++
 fs/f2fs/super.c                    | 119 ++++++++++++++++++++++++++++-
 include/linux/f2fs_fs.h            |   1 +
 13 files changed, 296 insertions(+), 13 deletions(-)
Documentation/filesystems/f2fs.txt +5 -0
index bde3e91e5372..e46c2147ddf8 100644
@@ -212,6 +212,11 @@ fsync_mode=%s          Control the policy of fsync. Currently supports "posix",
                        non-atomic files likewise "nobarrier" mount option.
 test_dummy_encryption  Enable dummy encryption, which provides a fake fscrypt
                        context. The fake fscrypt context is used by xfstests.
+checkpoint=%s          Set to "disable" to turn off checkpointing. Set to "enable"
+                       to reenable checkpointing. Is enabled by default. While
+                       disabled, any unmounting or unexpected shutdowns will cause
+                       the filesystem contents to appear as they did when the
+                       filesystem was mounted with that option.
 
 ================================================================================
 DEBUGFS ENTRIES
fs/f2fs/checkpoint.c +16 -0
index d312d2829d5a..01e0d8f5bbbe 100644
@@ -1214,6 +1214,11 @@ static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK))
 		__set_ckpt_flags(ckpt, CP_FSCK_FLAG);
 
+	if (is_sbi_flag_set(sbi, SBI_CP_DISABLED))
+		__set_ckpt_flags(ckpt, CP_DISABLED_FLAG);
+	else
+		__clear_ckpt_flags(ckpt, CP_DISABLED_FLAG);
+
 	/* 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);
@@ -1421,6 +1426,7 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	clear_sbi_flag(sbi, SBI_IS_DIRTY);
 	clear_sbi_flag(sbi, SBI_NEED_CP);
+	sbi->unusable_block_count = 0;
 	__set_cp_next_pack(sbi);
 
 	/*
@@ -1445,6 +1451,16 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	unsigned long long ckpt_ver;
 	int err = 0;
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) {
+		if (cpc->reason != CP_PAUSE) {
+			if (cpc->reason != CP_UMOUNT)
+				f2fs_msg(sbi->sb, KERN_WARNING,
+					"Wrong disabled checkpoint!");
+			return 0;
+		}
+		f2fs_msg(sbi->sb, KERN_WARNING,
+				"Start checkpoint disabled!");
+	}
 	mutex_lock(&sbi->cp_mutex);
 
 	if (!is_sbi_flag_set(sbi, SBI_IS_DIRTY) &&
fs/f2fs/data.c +13 -1
index cb3ebcae0398..dc66c1948d8c 100644
@@ -537,7 +537,8 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio)
 	if (fio->in_list)
 		goto next;
 out:
-	if (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN))
+	if (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN) ||
+				f2fs_is_checkpoint_ready(sbi))
 		__submit_merged_bio(io);
 	up_write(&io->io_rwsem);
 }
@@ -1695,6 +1696,10 @@ static inline bool check_inplace_update_policy(struct inode *inode,
 			is_inode_flag_set(inode, FI_NEED_IPU))
 		return true;
 
+	if (unlikely(fio && is_sbi_flag_set(sbi, SBI_CP_DISABLED) &&
+			!f2fs_is_checkpointed_data(sbi, fio->old_blkaddr)))
+		return true;
+
 	return false;
 }
 
@@ -1725,6 +1730,9 @@ bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio)
 			return true;
 		if (IS_ATOMIC_WRITTEN_PAGE(fio->page))
 			return true;
+		if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED) &&
+			f2fs_is_checkpointed_data(sbi, fio->old_blkaddr)))
+			return true;
 	}
 	return false;
 }
@@ -2345,6 +2353,10 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 
 	trace_f2fs_write_begin(inode, pos, len, flags);
 
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		goto fail;
+
 	if ((f2fs_is_atomic_file(inode) &&
 			!f2fs_available_free_memory(sbi, INMEM_PAGES)) ||
 			is_inode_flag_set(inode, FI_ATOMIC_REVOKE_REQUEST)) {
fs/f2fs/debug.c +2 -1
index d3c402183e3c..cb9e7679124b 100644
@@ -267,7 +267,8 @@ static int stat_show(struct seq_file *s, void *v)
 		seq_printf(s, "\n=====[ partition info(%pg). #%d, %s, CP: %s]=====\n",
 			si->sbi->sb->s_bdev, i++,
 			f2fs_readonly(si->sbi->sb) ? "RO": "RW",
-			f2fs_cp_error(si->sbi) ? "Error": "Good");
+			is_set_ckpt_flags(si->sbi, CP_DISABLED_FLAG) ?
+			"Disabled": (f2fs_cp_error(si->sbi) ? "Error": "Good"));
 		seq_printf(s, "[SB: 1] [CP: 2] [SIT: %d] [NAT: %d] ",
 			   si->sit_area_segs, si->nat_area_segs);
 		seq_printf(s, "[SSA: %d] [MAIN: %d",
fs/f2fs/f2fs.h +13 -2
index b32fd9f7da95..89ff070ee95a 100644
@@ -98,6 +98,7 @@ extern char *f2fs_fault_name[FAULT_MAX];
 #define F2FS_MOUNT_QUOTA		0x00400000
 #define F2FS_MOUNT_INLINE_XATTR_SIZE	0x00800000
 #define F2FS_MOUNT_RESERVE_ROOT		0x01000000
+#define F2FS_MOUNT_DISABLE_CHECKPOINT	0x02000000
 
 #define F2FS_OPTION(sbi)	((sbi)->mount_opt)
 #define clear_opt(sbi, option)	(F2FS_OPTION(sbi).opt &= ~F2FS_MOUNT_##option)
@@ -176,6 +177,7 @@ enum {
 #define	CP_RECOVERY	0x00000008
 #define	CP_DISCARD	0x00000010
 #define CP_TRIMMED	0x00000020
+#define CP_PAUSE	0x00000040
 
 #define MAX_DISCARD_BLOCKS(sbi)		BLKS_PER_SEC(sbi)
 #define DEF_MAX_DISCARD_REQUEST		8	/* issue 8 discards per round */
@@ -1089,6 +1091,7 @@ enum {
 	SBI_NEED_CP,				/* need to checkpoint */
 	SBI_IS_SHUTDOWN,			/* shutdown by ioctl */
 	SBI_IS_RECOVERED,			/* recovered orphan/data */
+	SBI_CP_DISABLED,			/* CP was disabled last mount */
 };
 
 enum {
@@ -1222,6 +1225,9 @@ struct f2fs_sb_info {
 	block_t reserved_blocks;		/* configurable reserved blocks */
 	block_t current_reserved_blocks;	/* current reserved blocks */
 
+	/* Additional tracking for no checkpoint mode */
+	block_t unusable_block_count;		/* # of blocks saved by last cp */
+
 	unsigned int nquota_files;		/* # of quota sysfile */
 
 	u32 s_next_generation;			/* for NFS support */
@@ -1729,7 +1735,8 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi,
 
 	if (!__allow_reserved_blocks(sbi, inode, true))
 		avail_user_block_count -= F2FS_OPTION(sbi).root_reserved_blocks;
-
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		avail_user_block_count -= sbi->unusable_block_count;
 	if (unlikely(sbi->total_valid_block_count > avail_user_block_count)) {
 		diff = sbi->total_valid_block_count - avail_user_block_count;
 		if (diff > *count)
@@ -1936,6 +1943,8 @@ static inline int inc_valid_node_count(struct f2fs_sb_info *sbi,
 
 	if (!__allow_reserved_blocks(sbi, inode, false))
 		valid_block_count += F2FS_OPTION(sbi).root_reserved_blocks;
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		valid_block_count += sbi->unusable_block_count;
 
 	if (unlikely(valid_block_count > sbi->user_block_count)) {
 		spin_unlock(&sbi->stat_lock);
@@ -2939,6 +2948,7 @@ void f2fs_stop_discard_thread(struct f2fs_sb_info *sbi);
 bool f2fs_wait_discard_bios(struct f2fs_sb_info *sbi);
 void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi,
 					struct cp_control *cpc);
+void f2fs_dirty_to_prefree(struct f2fs_sb_info *sbi);
 void f2fs_release_discard_addrs(struct f2fs_sb_info *sbi);
 int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra);
 void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi);
@@ -3469,7 +3479,8 @@ static inline bool f2fs_may_encrypt(struct inode *inode)
 static inline bool f2fs_force_buffered_io(struct inode *inode, int rw)
 {
 	return (f2fs_post_read_required(inode) ||
-			(rw == WRITE && test_opt(F2FS_I_SB(inode), LFS)) ||
+			(rw == WRITE && (test_opt(F2FS_I_SB(inode), LFS) ||
+			is_sbi_flag_set(F2FS_I_SB(inode), SBI_CP_DISABLED))) ||
 			F2FS_I_SB(inode)->s_ndevs);
 }
 
fs/f2fs/file.c +12 -0
index 357422a4c319..3f092bde0240 100644
@@ -146,6 +146,9 @@ static inline enum cp_reason_type need_do_checkpoint(struct inode *inode)
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	enum cp_reason_type cp_reason = CP_NO_NEEDED;
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		return CP_NO_NEEDED;
+
 	if (!S_ISREG(inode->i_mode))
 		cp_reason = CP_NON_REGULAR;
 	else if (inode->i_nlink != 1)
@@ -2158,6 +2161,12 @@ static int f2fs_ioc_write_checkpoint(struct file *filp, unsigned long arg)
 	if (f2fs_readonly(sbi->sb))
 		return -EROFS;
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) {
+		f2fs_msg(sbi->sb, KERN_INFO,
+			"Skipping Checkpoint. Checkpoints currently disabled.");
+		return -EINVAL;
+	}
+
 	ret = mnt_want_write_file(filp);
 	if (ret)
 		return ret;
@@ -2529,6 +2538,9 @@ static int f2fs_ioc_flush_device(struct file *filp, unsigned long arg)
 	if (f2fs_readonly(sbi->sb))
 		return -EROFS;
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		return -EINVAL;
+
 	if (copy_from_user(&range, (struct f2fs_flush_device __user *)arg,
 							sizeof(range)))
 		return -EFAULT;
fs/f2fs/gc.c +7 -2
index c051dc4ddf1a..5d2fa37472be 100644
@@ -362,6 +362,10 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 
 		if (sec_usage_check(sbi, secno))
 			goto next;
+		/* Don't touch checkpointed data */
+		if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED) &&
+					get_ckpt_valid_blocks(sbi, segno)))
+			goto next;
 		if (gc_type == BG_GC && test_bit(secno, dirty_i->victim_secmap))
 			goto next;
 
@@ -1169,7 +1173,8 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 		 * threshold, we can make them free by checkpoint. Then, we
 		 * secure free segments which doesn't need fggc any more.
 		 */
-		if (prefree_segments(sbi)) {
+		if (prefree_segments(sbi) &&
+				!is_sbi_flag_set(sbi, SBI_CP_DISABLED)) {
 			ret = f2fs_write_checkpoint(sbi, &cpc);
 			if (ret)
 				goto stop;
@@ -1221,7 +1226,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 			segno = NULL_SEGNO;
 			goto gc_more;
 		}
-		if (gc_type == FG_GC)
+		if (gc_type == FG_GC && !is_sbi_flag_set(sbi, SBI_CP_DISABLED))
 			ret = f2fs_write_checkpoint(sbi, &cpc);
 	}
 stop:
fs/f2fs/inode.c +5 -1
index 86e7333d60c1..4ee9d6c4b719 100644
@@ -607,6 +607,9 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 	if (!is_inode_flag_set(inode, FI_DIRTY_INODE))
 		return 0;
 
+	if (f2fs_is_checkpoint_ready(sbi))
+		return -ENOSPC;
+
 	/*
 	 * We need to balance fs here to prevent from producing dirty node pages
 	 * during the urgent cleaning time when runing out of free sections.
@@ -688,7 +691,8 @@ void f2fs_evict_inode(struct inode *inode)
 	stat_dec_inline_dir(inode);
 	stat_dec_inline_inode(inode);
 
-	if (likely(!is_set_ckpt_flags(sbi, CP_ERROR_FLAG)))
+	if (likely(!is_set_ckpt_flags(sbi, CP_ERROR_FLAG) &&
+				!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
 		f2fs_bug_on(sbi, is_inode_flag_set(inode, FI_DIRTY_INODE));
 	else
 		f2fs_inode_synced(inode);
fs/f2fs/namei.c +19 -0
index ded185baa9ae..a14632744a6a 100644
@@ -16,6 +16,7 @@
 
 #include "f2fs.h"
 #include "node.h"
+#include "segment.h"
 #include "xattr.h"
 #include "acl.h"
 #include <trace/events/f2fs.h>
@@ -269,6 +270,9 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = dquot_initialize(dir);
 	if (err)
@@ -315,6 +319,9 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = fscrypt_prepare_link(old_dentry, dir, dentry);
 	if (err)
@@ -561,6 +568,9 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = fscrypt_prepare_symlink(dir, symname, len, dir->i_sb->s_blocksize,
 				      &disk_link);
@@ -690,6 +700,9 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = dquot_initialize(dir);
 	if (err)
@@ -824,6 +837,9 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	if (is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
 			(!projid_eq(F2FS_I(new_dir)->i_projid,
@@ -1014,6 +1030,9 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	if ((is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
 			!projid_eq(F2FS_I(new_dir)->i_projid,
fs/f2fs/segment.c +71 -4

136 lines not on this page · raw message

fs/f2fs/segment.h +15 -0
index 086150028c6d..ab3465faddf1 100644
@@ -339,6 +339,12 @@ static inline unsigned int get_valid_blocks(struct f2fs_sb_info *sbi,
 		return get_seg_entry(sbi, segno)->valid_blocks;
 }
 
+static inline unsigned int get_ckpt_valid_blocks(struct f2fs_sb_info *sbi,
+				unsigned int segno)
+{
+	return get_seg_entry(sbi, segno)->ckpt_valid_blocks;
+}
+
 static inline void seg_info_from_raw_sit(struct seg_entry *se,
 					struct f2fs_sit_entry *rs)
 {
@@ -576,6 +582,15 @@ static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi,
 		reserved_sections(sbi) + needed);
 }
 
+static inline int f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
+{
+	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		return 0;
+	if (likely(!has_not_enough_free_secs(sbi, 0, 0)))
+		return 0;
+	return -ENOSPC;
+}
+
 static inline bool excess_prefree_segs(struct f2fs_sb_info *sbi)
 {
 	return prefree_segments(sbi) > SM_I(sbi)->rec_prefree_segments;
fs/f2fs/super.c +117 -2

221 lines not on this page · raw message

include/linux/f2fs_fs.h +1 -0
index 1d4b196291d6..9e95bbce5dd4 100644
@@ -115,6 +115,7 @@ struct f2fs_super_block {
 /*
  * For checkpoint
  */
+#define CP_DISABLED_FLAG		0x00001000
 #define CP_LARGE_NAT_BITMAP_FLAG	0x00000400
 #define CP_NOCRC_RECOVERY_FLAG	0x00000200
 #define CP_TRIMMED_FLAG		0x00000100
Signature · 1 line
2.17.0.441.gb46fe60e1d-goog
author maintainer
#1 lore raw

Note that, it requires "f2fs: return correct errno in f2fs_gc".

This adds a lightweight non-persistent snapshotting scheme to f2fs.

To use, mount with the option checkpoint=disable, and to return to normal operation, remount with checkpoint=enable. If the filesystem is shut down before remounting with checkpoint=enable, it will revert back to its apparent state when it was first mounted with checkpoint=disable. This is useful for situations where you wish to be able to roll back the state of the disk in case of some critical failure.

---
Change log from v1:
- fix fsync loop

 Documentation/filesystems/f2fs.txt |   5 ++
 fs/f2fs/checkpoint.c               |  16 ++++
 fs/f2fs/data.c                     |  14 +++-
 fs/f2fs/debug.c                    |   3 +-
 fs/f2fs/f2fs.h                     |  15 +++-
 fs/f2fs/file.c                     |  12 ++-
 fs/f2fs/gc.c                       |   9 ++-
 fs/f2fs/inode.c                    |   6 +-
 fs/f2fs/namei.c                    |  19 +++++
 fs/f2fs/segment.c                  |  75 +++++++++++++++++-
 fs/f2fs/segment.h                  |  15 ++++
 fs/f2fs/super.c                    | 118 ++++++++++++++++++++++++++++-
 include/linux/f2fs_fs.h            |   1 +
 13 files changed, 294 insertions(+), 14 deletions(-)
Documentation/filesystems/f2fs.txt +5 -0
index bde3e91e5372..e46c2147ddf8 100644
@@ -212,6 +212,11 @@ fsync_mode=%s          Control the policy of fsync. Currently supports "posix",
                        non-atomic files likewise "nobarrier" mount option.
 test_dummy_encryption  Enable dummy encryption, which provides a fake fscrypt
                        context. The fake fscrypt context is used by xfstests.
+checkpoint=%s          Set to "disable" to turn off checkpointing. Set to "enable"
+                       to reenable checkpointing. Is enabled by default. While
+                       disabled, any unmounting or unexpected shutdowns will cause
+                       the filesystem contents to appear as they did when the
+                       filesystem was mounted with that option.
 
 ================================================================================
 DEBUGFS ENTRIES
fs/f2fs/checkpoint.c +16 -0
index d312d2829d5a..01e0d8f5bbbe 100644
@@ -1214,6 +1214,11 @@ static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK))
 		__set_ckpt_flags(ckpt, CP_FSCK_FLAG);
 
+	if (is_sbi_flag_set(sbi, SBI_CP_DISABLED))
+		__set_ckpt_flags(ckpt, CP_DISABLED_FLAG);
+	else
+		__clear_ckpt_flags(ckpt, CP_DISABLED_FLAG);
+
 	/* 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);
@@ -1421,6 +1426,7 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	clear_sbi_flag(sbi, SBI_IS_DIRTY);
 	clear_sbi_flag(sbi, SBI_NEED_CP);
+	sbi->unusable_block_count = 0;
 	__set_cp_next_pack(sbi);
 
 	/*
@@ -1445,6 +1451,16 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	unsigned long long ckpt_ver;
 	int err = 0;
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) {
+		if (cpc->reason != CP_PAUSE) {
+			if (cpc->reason != CP_UMOUNT)
+				f2fs_msg(sbi->sb, KERN_WARNING,
+					"Wrong disabled checkpoint!");
+			return 0;
+		}
+		f2fs_msg(sbi->sb, KERN_WARNING,
+				"Start checkpoint disabled!");
+	}
 	mutex_lock(&sbi->cp_mutex);
 
 	if (!is_sbi_flag_set(sbi, SBI_IS_DIRTY) &&
fs/f2fs/data.c +13 -1
index cb3ebcae0398..dc66c1948d8c 100644
@@ -537,7 +537,8 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio)
 	if (fio->in_list)
 		goto next;
 out:
-	if (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN))
+	if (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN) ||
+				f2fs_is_checkpoint_ready(sbi))
 		__submit_merged_bio(io);
 	up_write(&io->io_rwsem);
 }
@@ -1695,6 +1696,10 @@ static inline bool check_inplace_update_policy(struct inode *inode,
 			is_inode_flag_set(inode, FI_NEED_IPU))
 		return true;
 
+	if (unlikely(fio && is_sbi_flag_set(sbi, SBI_CP_DISABLED) &&
+			!f2fs_is_checkpointed_data(sbi, fio->old_blkaddr)))
+		return true;
+
 	return false;
 }
 
@@ -1725,6 +1730,9 @@ bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio)
 			return true;
 		if (IS_ATOMIC_WRITTEN_PAGE(fio->page))
 			return true;
+		if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED) &&
+			f2fs_is_checkpointed_data(sbi, fio->old_blkaddr)))
+			return true;
 	}
 	return false;
 }
@@ -2345,6 +2353,10 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 
 	trace_f2fs_write_begin(inode, pos, len, flags);
 
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		goto fail;
+
 	if ((f2fs_is_atomic_file(inode) &&
 			!f2fs_available_free_memory(sbi, INMEM_PAGES)) ||
 			is_inode_flag_set(inode, FI_ATOMIC_REVOKE_REQUEST)) {
fs/f2fs/debug.c +2 -1
index d3c402183e3c..cb9e7679124b 100644
@@ -267,7 +267,8 @@ static int stat_show(struct seq_file *s, void *v)
 		seq_printf(s, "\n=====[ partition info(%pg). #%d, %s, CP: %s]=====\n",
 			si->sbi->sb->s_bdev, i++,
 			f2fs_readonly(si->sbi->sb) ? "RO": "RW",
-			f2fs_cp_error(si->sbi) ? "Error": "Good");
+			is_set_ckpt_flags(si->sbi, CP_DISABLED_FLAG) ?
+			"Disabled": (f2fs_cp_error(si->sbi) ? "Error": "Good"));
 		seq_printf(s, "[SB: 1] [CP: 2] [SIT: %d] [NAT: %d] ",
 			   si->sit_area_segs, si->nat_area_segs);
 		seq_printf(s, "[SSA: %d] [MAIN: %d",
fs/f2fs/f2fs.h +13 -2
index b32fd9f7da95..89ff070ee95a 100644
@@ -98,6 +98,7 @@ extern char *f2fs_fault_name[FAULT_MAX];
 #define F2FS_MOUNT_QUOTA		0x00400000
 #define F2FS_MOUNT_INLINE_XATTR_SIZE	0x00800000
 #define F2FS_MOUNT_RESERVE_ROOT		0x01000000
+#define F2FS_MOUNT_DISABLE_CHECKPOINT	0x02000000
 
 #define F2FS_OPTION(sbi)	((sbi)->mount_opt)
 #define clear_opt(sbi, option)	(F2FS_OPTION(sbi).opt &= ~F2FS_MOUNT_##option)
@@ -176,6 +177,7 @@ enum {
 #define	CP_RECOVERY	0x00000008
 #define	CP_DISCARD	0x00000010
 #define CP_TRIMMED	0x00000020
+#define CP_PAUSE	0x00000040
 
 #define MAX_DISCARD_BLOCKS(sbi)		BLKS_PER_SEC(sbi)
 #define DEF_MAX_DISCARD_REQUEST		8	/* issue 8 discards per round */
@@ -1089,6 +1091,7 @@ enum {
 	SBI_NEED_CP,				/* need to checkpoint */
 	SBI_IS_SHUTDOWN,			/* shutdown by ioctl */
 	SBI_IS_RECOVERED,			/* recovered orphan/data */
+	SBI_CP_DISABLED,			/* CP was disabled last mount */
 };
 
 enum {
@@ -1222,6 +1225,9 @@ struct f2fs_sb_info {
 	block_t reserved_blocks;		/* configurable reserved blocks */
 	block_t current_reserved_blocks;	/* current reserved blocks */
 
+	/* Additional tracking for no checkpoint mode */
+	block_t unusable_block_count;		/* # of blocks saved by last cp */
+
 	unsigned int nquota_files;		/* # of quota sysfile */
 
 	u32 s_next_generation;			/* for NFS support */
@@ -1729,7 +1735,8 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi,
 
 	if (!__allow_reserved_blocks(sbi, inode, true))
 		avail_user_block_count -= F2FS_OPTION(sbi).root_reserved_blocks;
-
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		avail_user_block_count -= sbi->unusable_block_count;
 	if (unlikely(sbi->total_valid_block_count > avail_user_block_count)) {
 		diff = sbi->total_valid_block_count - avail_user_block_count;
 		if (diff > *count)
@@ -1936,6 +1943,8 @@ static inline int inc_valid_node_count(struct f2fs_sb_info *sbi,
 
 	if (!__allow_reserved_blocks(sbi, inode, false))
 		valid_block_count += F2FS_OPTION(sbi).root_reserved_blocks;
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		valid_block_count += sbi->unusable_block_count;
 
 	if (unlikely(valid_block_count > sbi->user_block_count)) {
 		spin_unlock(&sbi->stat_lock);
@@ -2939,6 +2948,7 @@ void f2fs_stop_discard_thread(struct f2fs_sb_info *sbi);
 bool f2fs_wait_discard_bios(struct f2fs_sb_info *sbi);
 void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi,
 					struct cp_control *cpc);
+void f2fs_dirty_to_prefree(struct f2fs_sb_info *sbi);
 void f2fs_release_discard_addrs(struct f2fs_sb_info *sbi);
 int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra);
 void f2fs_allocate_new_segments(struct f2fs_sb_info *sbi);
@@ -3469,7 +3479,8 @@ static inline bool f2fs_may_encrypt(struct inode *inode)
 static inline bool f2fs_force_buffered_io(struct inode *inode, int rw)
 {
 	return (f2fs_post_read_required(inode) ||
-			(rw == WRITE && test_opt(F2FS_I_SB(inode), LFS)) ||
+			(rw == WRITE && (test_opt(F2FS_I_SB(inode), LFS) ||
+			is_sbi_flag_set(F2FS_I_SB(inode), SBI_CP_DISABLED))) ||
 			F2FS_I_SB(inode)->s_ndevs);
 }
 
fs/f2fs/file.c +11 -1
index 357422a4c319..d68a1b00e3fd 100644
@@ -211,7 +211,8 @@ static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
 	};
 	unsigned int seq_id = 0;
 
-	if (unlikely(f2fs_readonly(inode->i_sb)))
+	if (unlikely(f2fs_readonly(inode->i_sb) ||
+				is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
 		return 0;
 
 	trace_f2fs_sync_file_enter(inode);
@@ -2158,6 +2159,12 @@ static int f2fs_ioc_write_checkpoint(struct file *filp, unsigned long arg)
 	if (f2fs_readonly(sbi->sb))
 		return -EROFS;
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) {
+		f2fs_msg(sbi->sb, KERN_INFO,
+			"Skipping Checkpoint. Checkpoints currently disabled.");
+		return -EINVAL;
+	}
+
 	ret = mnt_want_write_file(filp);
 	if (ret)
 		return ret;
@@ -2529,6 +2536,9 @@ static int f2fs_ioc_flush_device(struct file *filp, unsigned long arg)
 	if (f2fs_readonly(sbi->sb))
 		return -EROFS;
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		return -EINVAL;
+
 	if (copy_from_user(&range, (struct f2fs_flush_device __user *)arg,
 							sizeof(range)))
 		return -EFAULT;
fs/f2fs/gc.c +7 -2
index c051dc4ddf1a..5d2fa37472be 100644
@@ -362,6 +362,10 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 
 		if (sec_usage_check(sbi, secno))
 			goto next;
+		/* Don't touch checkpointed data */
+		if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED) &&
+					get_ckpt_valid_blocks(sbi, segno)))
+			goto next;
 		if (gc_type == BG_GC && test_bit(secno, dirty_i->victim_secmap))
 			goto next;
 
@@ -1169,7 +1173,8 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 		 * threshold, we can make them free by checkpoint. Then, we
 		 * secure free segments which doesn't need fggc any more.
 		 */
-		if (prefree_segments(sbi)) {
+		if (prefree_segments(sbi) &&
+				!is_sbi_flag_set(sbi, SBI_CP_DISABLED)) {
 			ret = f2fs_write_checkpoint(sbi, &cpc);
 			if (ret)
 				goto stop;
@@ -1221,7 +1226,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 			segno = NULL_SEGNO;
 			goto gc_more;
 		}
-		if (gc_type == FG_GC)
+		if (gc_type == FG_GC && !is_sbi_flag_set(sbi, SBI_CP_DISABLED))
 			ret = f2fs_write_checkpoint(sbi, &cpc);
 	}
 stop:
fs/f2fs/inode.c +5 -1
index 86e7333d60c1..4ee9d6c4b719 100644
@@ -607,6 +607,9 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 	if (!is_inode_flag_set(inode, FI_DIRTY_INODE))
 		return 0;
 
+	if (f2fs_is_checkpoint_ready(sbi))
+		return -ENOSPC;
+
 	/*
 	 * We need to balance fs here to prevent from producing dirty node pages
 	 * during the urgent cleaning time when runing out of free sections.
@@ -688,7 +691,8 @@ void f2fs_evict_inode(struct inode *inode)
 	stat_dec_inline_dir(inode);
 	stat_dec_inline_inode(inode);
 
-	if (likely(!is_set_ckpt_flags(sbi, CP_ERROR_FLAG)))
+	if (likely(!is_set_ckpt_flags(sbi, CP_ERROR_FLAG) &&
+				!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
 		f2fs_bug_on(sbi, is_inode_flag_set(inode, FI_DIRTY_INODE));
 	else
 		f2fs_inode_synced(inode);
fs/f2fs/namei.c +19 -0
index ded185baa9ae..a14632744a6a 100644
@@ -16,6 +16,7 @@
 
 #include "f2fs.h"
 #include "node.h"
+#include "segment.h"
 #include "xattr.h"
 #include "acl.h"
 #include <trace/events/f2fs.h>
@@ -269,6 +270,9 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = dquot_initialize(dir);
 	if (err)
@@ -315,6 +319,9 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = fscrypt_prepare_link(old_dentry, dir, dentry);
 	if (err)
@@ -561,6 +568,9 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = fscrypt_prepare_symlink(dir, symname, len, dir->i_sb->s_blocksize,
 				      &disk_link);
@@ -690,6 +700,9 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	err = dquot_initialize(dir);
 	if (err)
@@ -824,6 +837,9 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	if (is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
 			(!projid_eq(F2FS_I(new_dir)->i_projid,
@@ -1014,6 +1030,9 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	if (unlikely(f2fs_cp_error(sbi)))
 		return -EIO;
+	err = f2fs_is_checkpoint_ready(sbi);
+	if (err)
+		return err;
 
 	if ((is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
 			!projid_eq(F2FS_I(new_dir)->i_projid,
fs/f2fs/segment.c +71 -4

136 lines not on this page · raw message

fs/f2fs/segment.h +15 -0
index 086150028c6d..ab3465faddf1 100644
@@ -339,6 +339,12 @@ static inline unsigned int get_valid_blocks(struct f2fs_sb_info *sbi,
 		return get_seg_entry(sbi, segno)->valid_blocks;
 }
 
+static inline unsigned int get_ckpt_valid_blocks(struct f2fs_sb_info *sbi,
+				unsigned int segno)
+{
+	return get_seg_entry(sbi, segno)->ckpt_valid_blocks;
+}
+
 static inline void seg_info_from_raw_sit(struct seg_entry *se,
 					struct f2fs_sit_entry *rs)
 {
@@ -576,6 +582,15 @@ static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi,
 		reserved_sections(sbi) + needed);
 }
 
+static inline int f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
+{
+	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
+		return 0;
+	if (likely(!has_not_enough_free_secs(sbi, 0, 0)))
+		return 0;
+	return -ENOSPC;
+}
+
 static inline bool excess_prefree_segs(struct f2fs_sb_info *sbi)
 {
 	return prefree_segments(sbi) > SM_I(sbi)->rec_prefree_segments;
fs/f2fs/super.c +116 -2

214 lines not on this page · raw message

include/linux/f2fs_fs.h +1 -0
index 1d4b196291d6..9e95bbce5dd4 100644
@@ -115,6 +115,7 @@ struct f2fs_super_block {
 /*
  * For checkpoint
  */
+#define CP_DISABLED_FLAG		0x00001000
 #define CP_LARGE_NAT_BITMAP_FLAG	0x00000400
 #define CP_NOCRC_RECOVERY_FLAG	0x00000200
 #define CP_TRIMMED_FLAG		0x00000100
Signature · 1 line
2.19.0.605.g01d371f741-goog