f2fs: checkpoint disabling

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. Signed-off-by: Daniel Rosenberg <[email protected]> [Jaegeuk Kim: use SB_RDONLY instead of MS_RDONLY] Signed-off-by: Jaegeuk Kim <[email protected]>
Ruega · Oct 30, 2018 · 99efe50 · 99efe50
1 parent bb423cf
commit 99efe50
Show file tree

Hide file tree

Showing 13 changed files with 323 additions and 13 deletions.
diff --git a/Documentation/filesystems/f2fs.txt b/Documentation/filesystems/f2fs.txt
@@ -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

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
@@ -1211,6 +1211,11 @@ 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);
@@ -1418,6 +1423,7 @@ 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);
 
 	/*
@@ -1442,6 +1448,12 @@ 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)
+			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) &&

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
@@ -534,7 +534,8 @@ 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);
 }
@@ -1699,6 +1700,10 @@ 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;
 }
 
@@ -1729,6 +1734,9 @@ 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;
 }
@@ -2349,6 +2357,10 @@ 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)) {

diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c
@@ -272,7 +272,8 @@ 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",

diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
@@ -101,6 +101,7 @@ 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)
@@ -246,6 +247,7 @@ 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 */
@@ -255,6 +257,7 @@ enum {
 #define DEF_DISCARD_URGENT_UTIL		80	/* do more discard over 80% */
 #define DEF_CP_INTERVAL			60	/* 60 secs */
 #define DEF_IDLE_INTERVAL		5	/* 5 secs */
+#define DEF_DISABLE_INTERVAL		5	/* 5 secs */
 
 struct cp_control {
 	int reason;
@@ -1157,13 +1160,15 @@ 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 {
 	CP_TIME,
 	REQ_TIME,
 	DISCARD_TIME,
 	GC_TIME,
+	DISABLE_TIME,
 	MAX_TIME,
 };
 
@@ -1290,6 +1295,9 @@ 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 */
@@ -1800,7 +1808,8 @@ 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)
@@ -2007,6 +2016,8 @@ 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);
@@ -3030,6 +3041,8 @@ 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);
+int f2fs_disable_cp_again(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);
@@ -3617,6 +3630,9 @@ static inline bool f2fs_force_buffered_io(struct inode *inode,
 	if (test_opt(sbi, LFS) && (rw == WRITE) &&
 				block_unaligned_IO(inode, iocb, iter))
 		return true;
+	if (is_sbi_flag_set(F2FS_I_SB(inode), SBI_CP_DISABLED))
+		return true;
+
 	return false;
 }
 

diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
@@ -213,7 +213,8 @@ 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);
@@ -2157,6 +2158,12 @@ 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;
@@ -2528,6 +2535,9 @@ 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;

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
@@ -370,6 +370,10 @@ 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;
 
@@ -1189,7 +1193,8 @@ 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;
@@ -1241,7 +1246,7 @@ 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:

diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
@@ -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);

diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
@@ -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)
@@ -316,6 +320,9 @@ 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)
@@ -559,6 +566,9 @@ 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)
@@ -825,6 +838,9 @@ 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,
@@ -1015,6 +1031,9 @@ 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,