1FREEZE_SUPER(9) The Linux VFS FREEZE_SUPER(9)
2
3
4
6 freeze_super - lock the filesystem and force it into a consistent state
7
9 int freeze_super(struct super_block * sb);
10
12 sb
13 the super to lock
14
16 Syncs the super to make sure the filesystem is consistent and calls the
17 fs's freeze_fs. Subsequent calls to this without first thawing the fs
18 will return -EBUSY.
19
20 During this function, sb->s_writers.frozen goes through these values:
21
23 File system is normal, all writes progress as usual.
24
26 The file system is in the process of being frozen. New writes should be
27 blocked, though page faults are still allowed. We wait for all writes
28 to complete and then proceed to the next stage.
29
31 Freezing continues. Now also page faults are blocked but internal fs
32 threads can still modify the filesystem (although they should not dirty
33 new pages or inodes), writeback can run etc. After waiting for all
34 running page faults we sync the filesystem which will clean all dirty
35 pages and inodes (no new dirty pages or inodes can be created when sync
36 is running).
37
39 The file system is frozen. Now all internal sources of fs modification
40 are blocked (e.g. XFS preallocation truncation on inode reclaim). This
41 is usually implemented by blocking new transactions for filesystems
42 that have them and need this additional guard. After all internal
43 writers are finished we call ->freeze_fs to finish filesystem freezing.
44 Then we transition to SB_FREEZE_COMPLETE state. This state is mostly
45 auxiliary for filesystems to verify they do not modify frozen fs.
46
47 sb->s_writers.frozen is protected by sb->s_umount.
48
50Kernel Hackers Manual 3.10 June 2019 FREEZE_SUPER(9)