ThreadA ThreadB
- f2fs_ioc_start_atomic_write
- write
- f2fs_ioc_commit_atomic_write
- f2fs_commit_inmem_pages
- f2fs_drop_inmem_pages
- f2fs_drop_inmem_pages
- __revoke_inmem_pages
- f2fs_vm_page_mkwrite
- set_page_dirty
- tag ATOMIC_WRITTEN_PAGE and add page
to inmem_pages list
- clear_inode_flag(FI_ATOMIC_FILE)
- f2fs_vm_page_mkwrite
- set_page_dirty
- f2fs_update_dirty_page
- f2fs_trace_pid
- tag inmem page private to pid
- truncate
- f2fs_invalidate_page
- set page->mapping to NULL
then it will cause panic once we
access page->mapping
The root cause is we missed to keep isolation of atomic write in the case
of commit_atomic_write vs mkwrite, let commit_atomic_write helds i_mmap_sem
lock to avoid this issue.
Signed-off-by: Chao Yu <[email protected]>
---
fs/f2fs/file.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 22a0101538c0..1ff5fc10e1fa 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -2094,10 +2094,12 @@ static int f2fs_ioc_commit_atomic_write(struct file *filp)
goto err_out;
}
+ down_write(&F2FS_I(inode)->i_mmap_sem);
+
if (f2fs_is_atomic_file(inode)) {
ret = f2fs_commit_inmem_pages(inode);
if (ret)
- goto err_out;
+ goto up_write;
ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 0, true);
if (!ret)
@@ -2105,6 +2107,8 @@ static int f2fs_ioc_commit_atomic_write(struct file *filp)
} else {
ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 1, false);
}
+up_write:
+ up_write(&F2FS_I(inode)->i_mmap_sem);
err_out:
if (is_inode_flag_set(inode, FI_ATOMIC_REVOKE_REQUEST)) {
clear_inode_flag(inode, FI_ATOMIC_REVOKE_REQUEST);
--
2.29.2