hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICJRGE -------------------------------- A race condition can occur in cifs_oplock_break() leading to a use-after-free of the cinode structure when unmounting: cifs_oplock_break() _cifsFileInfo_put(cfile) cifsFileInfo_put_final() cifs_sb_deactive() [last ref, start releasing sb] kill_sb() kill_anon_super() generic_shutdown_super() evict_inodes() dispose_list() evict() destroy_inode() call_rcu(&inode->i_rcu, i_callback) spin_lock(&cinode->open_file_lock) <- OK [later] i_callback() cifs_free_inode() kmem_cache_free(cinode) spin_unlock(&cinode->open_file_lock) <- UAF cifs_done_oplock_break(cinode) <- UAF The issue occurs when umount has already released its reference to the superblock. When _cifsFileInfo_put() calls cifs_sb_deactive(), this releases the last reference, triggering the immediate cleanup of all inodes under RCU. However, cifs_oplock_break() continues to access the cinode after this point, resulting in use-after-free. Fix this by holding an extra reference to the superblock during the entire oplock break operation. This ensures that the superblock and its inodes remain valid until the oplock break completes. Link: https://bugzilla.kernel.org/show_bug.cgi?id=220309 Fixes: b98749cac4a6 ("CIFS: keep FileInfo handle live during oplock break") Signed-off-by: Wang Zhaolong <wangzhaolong@huaweicloud.com> --- fs/smb/client/file.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index e19a4927493c..00a9c9c3db37 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -5022,20 +5022,27 @@ static int cifs_launder_folio(struct folio *folio) void cifs_oplock_break(struct work_struct *work) { struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo, oplock_break); struct inode *inode = d_inode(cfile->dentry); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct super_block *sb = inode->i_sb; + struct cifs_sb_info *cifs_sb = CIFS_SB(sb); struct cifsInodeInfo *cinode = CIFS_I(inode); struct cifs_tcon *tcon; struct TCP_Server_Info *server; struct tcon_link *tlink; int rc = 0; bool purge_cache = false, oplock_break_cancelled; __u64 persistent_fid, volatile_fid; __u16 net_fid; + /* + * Hold a reference to the superblock to prevent it and its inodes from + * being freed while we are accessing cinode. Otherwise, _cifsFileInfo_put() + * may release the last reference to the sb and trigger inode eviction. + */ + cifs_sb_active(sb); wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS, TASK_UNINTERRUPTIBLE); tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) @@ -5104,10 +5111,11 @@ void cifs_oplock_break(struct work_struct *work) spin_unlock(&cinode->open_file_lock); cifs_put_tlink(tlink); out: cifs_done_oplock_break(cinode); + cifs_sb_deactive(sb); } /* * The presence of cifs_direct_io() in the address space ops vector * allowes open() O_DIRECT flags which would have failed otherwise. -- 2.34.3