]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
do_umount(): add missing barrier before refcount checks in sync case
authorAl Viro <viro@zeniv.linux.org.uk>
Tue, 29 Apr 2025 03:56:14 +0000 (23:56 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sun, 18 May 2025 06:24:10 +0000 (08:24 +0200)
commit00f0dd1a0166e1ed8adeb50a6e345b4fd7878431
tree7df59a5304cba619272af4e24af63ff29837cdfc
parenteba09f42393abe8c70e78dd643166999b3144b75
do_umount(): add missing barrier before refcount checks in sync case

[ Upstream commit 65781e19dcfcb4aed1167d87a3ffcc2a0c071d47 ]

do_umount() analogue of the race fixed in 119e1ef80ecf "fix
__legitimize_mnt()/mntput() race".  Here we want to make sure that
if __legitimize_mnt() doesn't notice our lock_mount_hash(), we will
notice their refcount increment.  Harder to hit than mntput_no_expire()
one, fortunately, and consequences are milder (sync umount acting
like umount -l on a rare race with RCU pathwalk hitting at just the
wrong time instead of use-after-free galore mntput_no_expire()
counterpart used to be hit).  Still a bug...

Fixes: 48a066e72d97 ("RCU'd vfsmounts")
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
fs/namespace.c