ceph: drop messages from MDS when unmounting

When unmounting all the dirty buffers will be flushed and after
the last osd request is finished the last reference of the i_count
will be released. Then it will flush the dirty cap/snap to MDSs,
and the unmounting won't wait the possible acks, which will ihold
the inodes when updating the metadata locally but makes no sense
any more, of this. This will make the evict_inodes() to skip these
inodes.

If encrypt is enabled the kernel generate a warning when removing
the encrypt keys when the skipped inodes still hold the keyring:

WARNING: CPU: 4 PID: 168846 at fs/crypto/keyring.c:242 fscrypt_destroy_keyring+0x7e/0xd0
CPU: 4 PID: 168846 Comm: umount Tainted: G S  6.1.0-rc5-ceph-g72ead199864c #1
Hardware name: Supermicro SYS-5018R-WR/X10SRW-F, BIOS 2.0 12/17/2015
RIP: 0010:fscrypt_destroy_keyring+0x7e/0xd0
RSP: 0018:ffffc9000b277e28 EFLAGS: 00010202
RAX: 0000000000000002 RBX: ffff88810d52ac00 RCX: ffff88810b56aa00
RDX: 0000000080000000 RSI: ffffffff822f3a09 RDI: ffff888108f59000
RBP: ffff8881d394fb88 R08: 0000000000000028 R09: 0000000000000000
R10: 0000000000000001 R11: 11ff4fe6834fcd91 R12: ffff8881d394fc40
R13: ffff888108f59000 R14: ffff8881d394f800 R15: 0000000000000000
FS:  00007fd83f6f1080(0000) GS:ffff88885fd00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f918d417000 CR3: 000000017f89a005 CR4: 00000000003706e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
<TASK>
generic_shutdown_super+0x47/0x120
kill_anon_super+0x14/0x30
ceph_kill_sb+0x36/0x90 [ceph]
deactivate_locked_super+0x29/0x60
cleanup_mnt+0xb8/0x140
task_work_run+0x67/0xb0
exit_to_user_mode_prepare+0x23d/0x240
syscall_exit_to_user_mode+0x25/0x60
do_syscall_64+0x40/0x80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7fd83dc39e9b

Later the kernel will crash when iput() the inodes and dereferencing
the "sb->s_master_keys", which has been released by the
generic_shutdown_super().

Link: https://tracker.ceph.com/issues/59162
Signed-off-by: Xiubo Li <xiubli@redhat.com>
Reviewed-and-tested-by: Luís Henriques <lhenriques@suse.de>
Reviewed-by: Milind Changire <mchangir@redhat.com>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>

1 files changed, 70 insertions, 5 deletions

diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 75dd1b6b3d01..1c14d87ed871 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -1462,25 +1462,90 @@ nomem:
 	return -ENOMEM;
 }
 
+/*
+ * Return true if it successfully increases the blocker counter,
+ * or false if the mdsc is in stopping and flushed state.
+ */
+static bool __inc_stopping_blocker(struct ceph_mds_client *mdsc)
+{
+	spin_lock(&mdsc->stopping_lock);
+	if (mdsc->stopping >= CEPH_MDSC_STOPPING_FLUSHING) {
+		spin_unlock(&mdsc->stopping_lock);
+		return false;
+	}
+	atomic_inc(&mdsc->stopping_blockers);
+	spin_unlock(&mdsc->stopping_lock);
+	return true;
+}
+
+static void __dec_stopping_blocker(struct ceph_mds_client *mdsc)
+{
+	spin_lock(&mdsc->stopping_lock);
+	if (!atomic_dec_return(&mdsc->stopping_blockers) &&
+	    mdsc->stopping >= CEPH_MDSC_STOPPING_FLUSHING)
+		complete_all(&mdsc->stopping_waiter);
+	spin_unlock(&mdsc->stopping_lock);
+}
+
+/* For metadata IO requests */
+bool ceph_inc_mds_stopping_blocker(struct ceph_mds_client *mdsc,
+				   struct ceph_mds_session *session)
+{
+	mutex_lock(&session->s_mutex);
+	inc_session_sequence(session);
+	mutex_unlock(&session->s_mutex);
+
+	return __inc_stopping_blocker(mdsc);
+}
+
+void ceph_dec_mds_stopping_blocker(struct ceph_mds_client *mdsc)
+{
+	__dec_stopping_blocker(mdsc);
+}
+
 static void ceph_kill_sb(struct super_block *s)
 {
 	struct ceph_fs_client *fsc = ceph_sb_to_client(s);
+	struct ceph_mds_client *mdsc = fsc->mdsc;
+	bool wait;
 
 	dout("kill_sb %p\n", s);
 
-	ceph_mdsc_pre_umount(fsc->mdsc);
+	ceph_mdsc_pre_umount(mdsc);
 	flush_fs_workqueues(fsc);
 
 	/*
 	 * Though the kill_anon_super() will finally trigger the
-	 * sync_filesystem() anyway, we still need to do it here
-	 * and then bump the stage of shutdown to stop the work
-	 * queue as earlier as possible.
+	 * sync_filesystem() anyway, we still need to do it here and
+	 * then bump the stage of shutdown. This will allow us to
+	 * drop any further message, which will increase the inodes'
+	 * i_count reference counters but makes no sense any more,
+	 * from MDSs.
+	 *
+	 * Without this when evicting the inodes it may fail in the
+	 * kill_anon_super(), which will trigger a warning when
+	 * destroying the fscrypt keyring and then possibly trigger
+	 * a further crash in ceph module when the iput() tries to
+	 * evict the inodes later.
 	 */
 	sync_filesystem(s);
 
-	fsc->mdsc->stopping = CEPH_MDSC_STOPPING_FLUSHED;
+	spin_lock(&mdsc->stopping_lock);
+	mdsc->stopping = CEPH_MDSC_STOPPING_FLUSHING;
+	wait = !!atomic_read(&mdsc->stopping_blockers);
+	spin_unlock(&mdsc->stopping_lock);
+
+	if (wait && atomic_read(&mdsc->stopping_blockers)) {
+		long timeleft = wait_for_completion_killable_timeout(
+					&mdsc->stopping_waiter,
+					fsc->client->options->mount_timeout);
+		if (!timeleft) /* timed out */
+			pr_warn("umount timed out, %ld\n", timeleft);
+		else if (timeleft < 0) /* killed */
+			pr_warn("umount was killed, %ld\n", timeleft);
+	}
 
+	mdsc->stopping = CEPH_MDSC_STOPPING_FLUSHED;
 	kill_anon_super(s);
 
 	fsc->client->extra_mon_dispatch = NULL;