fs: fix NULL dereference due to data race in prepend_path()

Fix data race in prepend_path() with re-reading mnt->mnt_ns twice
without holding the lock.

is_mounted() does check for NULL, but is_anon_ns(mnt->mnt_ns) might
re-read the pointer again which could be NULL already, if in between
reads one of kern_unmount()/kern_unmount_array()/umount_tree() sets
mnt->mnt_ns to NULL.

This is seen in production with the following stack trace:

  BUG: kernel NULL pointer dereference, address: 0000000000000048
  ...
  RIP: 0010:prepend_path.isra.4+0x1ce/0x2e0
  Call Trace:
    d_path+0xe6/0x150
    proc_pid_readlink+0x8f/0x100
    vfs_readlink+0xf8/0x110
    do_readlinkat+0xfd/0x120
    __x64_sys_readlinkat+0x1a/0x20
    do_syscall_64+0x42/0x110
    entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fixes: f2683bd8d5 ("[PATCH] fix d_absolute_path() interplay with fsmount()")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Andrii Nakryiko 2020-10-14 13:45:28 -07:00 committed by Linus Torvalds
parent 4da9af0014
commit 09cad07547

View File

@ -102,6 +102,8 @@ static int prepend_path(const struct path *path,
if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) { if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) {
struct mount *parent = READ_ONCE(mnt->mnt_parent); struct mount *parent = READ_ONCE(mnt->mnt_parent);
struct mnt_namespace *mnt_ns;
/* Escaped? */ /* Escaped? */
if (dentry != vfsmnt->mnt_root) { if (dentry != vfsmnt->mnt_root) {
bptr = *buffer; bptr = *buffer;
@ -116,7 +118,9 @@ static int prepend_path(const struct path *path,
vfsmnt = &mnt->mnt; vfsmnt = &mnt->mnt;
continue; continue;
} }
if (is_mounted(vfsmnt) && !is_anon_ns(mnt->mnt_ns)) mnt_ns = READ_ONCE(mnt->mnt_ns);
/* open-coded is_mounted() to use local mnt_ns */
if (!IS_ERR_OR_NULL(mnt_ns) && !is_anon_ns(mnt_ns))
error = 1; // absolute root error = 1; // absolute root
else else
error = 2; // detached or not attached yet error = 2; // detached or not attached yet