net/smc: fix kernel panic caused by race of smc_sock

A crash occurs when smc_cdc_tx_handler() tries to access smc_sock
but smc_release() has already freed it.

[ 4570.695099] BUG: unable to handle page fault for address: 000000002eae9e88
[ 4570.696048] #PF: supervisor write access in kernel mode
[ 4570.696728] #PF: error_code(0x0002) - not-present page
[ 4570.697401] PGD 0 P4D 0
[ 4570.697716] Oops: 0002 [#1] PREEMPT SMP NOPTI
[ 4570.698228] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.16.0-rc4+ #111
[ 4570.699013] Hardware name: Alibaba Cloud Alibaba Cloud ECS, BIOS 8c24b4c 04/0
[ 4570.699933] RIP: 0010:_raw_spin_lock+0x1a/0x30
<...>
[ 4570.711446] Call Trace:
[ 4570.711746]  <IRQ>
[ 4570.711992]  smc_cdc_tx_handler+0x41/0xc0
[ 4570.712470]  smc_wr_tx_tasklet_fn+0x213/0x560
[ 4570.712981]  ? smc_cdc_tx_dismisser+0x10/0x10
[ 4570.713489]  tasklet_action_common.isra.17+0x66/0x140
[ 4570.714083]  __do_softirq+0x123/0x2f4
[ 4570.714521]  irq_exit_rcu+0xc4/0xf0
[ 4570.714934]  common_interrupt+0xba/0xe0

Though smc_cdc_tx_handler() checked the existence of smc connection,
smc_release() may have already dismissed and released the smc socket
before smc_cdc_tx_handler() further visits it.

smc_cdc_tx_handler()           |smc_release()
if (!conn)                     |
                               |
                               |smc_cdc_tx_dismiss_slots()
                               |      smc_cdc_tx_dismisser()
                               |
                               |sock_put(&smc->sk) <- last sock_put,
                               |                      smc_sock freed
bh_lock_sock(&smc->sk) (panic) |

To make sure we won't receive any CDC messages after we free the
smc_sock, add a refcount on the smc_connection for inflight CDC
message(posted to the QP but haven't received related CQE), and
don't release the smc_connection until all the inflight CDC messages
haven been done, for both success or failed ones.

Using refcount on CDC messages brings another problem: when the link
is going to be destroyed, smcr_link_clear() will reset the QP, which
then remove all the pending CQEs related to the QP in the CQ. To make
sure all the CQEs will always come back so the refcount on the
smc_connection can always reach 0, smc_ib_modify_qp_reset() was replaced
by smc_ib_modify_qp_error().
And remove the timeout in smc_wr_tx_wait_no_pending_sends() since we
need to wait for all pending WQEs done, or we may encounter use-after-
free when handling CQEs.

For IB device removal routine, we need to wait for all the QPs on that
device been destroyed before we can destroy CQs on the device, or
the refcount on smc_connection won't reach 0 and smc_sock cannot be
released.

Fixes: 5f08318f617b ("smc: connection data control (CDC)")
Reported-by: Wen Gu <guwen@linux.alibaba.com>
Signed-off-by: Dust Li <dust.li@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 24 insertions, 28 deletions

diff --git a/net/smc/smc_cdc.c b/net/smc/smc_cdc.c
index 99acd337ba90..84c8a4374fdd 100644
--- a/net/smc/smc_cdc.c
+++ b/net/smc/smc_cdc.c
@@ -31,10 +31,6 @@ static void smc_cdc_tx_handler(struct smc_wr_tx_pend_priv *pnd_snd,
 	struct smc_sock *smc;
 	int diff;
 
-	if (!conn)
-		/* already dismissed */
-		return;
-
 	smc = container_of(conn, struct smc_sock, conn);
 	bh_lock_sock(&smc->sk);
 	if (!wc_status) {
@@ -51,6 +47,12 @@ static void smc_cdc_tx_handler(struct smc_wr_tx_pend_priv *pnd_snd,
 			      conn);
 		conn->tx_cdc_seq_fin = cdcpend->ctrl_seq;
 	}
+
+	if (atomic_dec_and_test(&conn->cdc_pend_tx_wr) &&
+	    unlikely(wq_has_sleeper(&conn->cdc_pend_tx_wq)))
+		wake_up(&conn->cdc_pend_tx_wq);
+	WARN_ON(atomic_read(&conn->cdc_pend_tx_wr) < 0);
+
 	smc_tx_sndbuf_nonfull(smc);
 	bh_unlock_sock(&smc->sk);
 }
@@ -107,6 +109,10 @@ int smc_cdc_msg_send(struct smc_connection *conn,
 	conn->tx_cdc_seq++;
 	conn->local_tx_ctrl.seqno = conn->tx_cdc_seq;
 	smc_host_msg_to_cdc((struct smc_cdc_msg *)wr_buf, conn, &cfed);
+
+	atomic_inc(&conn->cdc_pend_tx_wr);
+	smp_mb__after_atomic(); /* Make sure cdc_pend_tx_wr added before post */
+
 	rc = smc_wr_tx_send(link, (struct smc_wr_tx_pend_priv *)pend);
 	if (!rc) {
 		smc_curs_copy(&conn->rx_curs_confirmed, &cfed, conn);
@@ -114,6 +120,7 @@ int smc_cdc_msg_send(struct smc_connection *conn,
 	} else {
 		conn->tx_cdc_seq--;
 		conn->local_tx_ctrl.seqno = conn->tx_cdc_seq;
+		atomic_dec(&conn->cdc_pend_tx_wr);
 	}
 
 	return rc;
@@ -136,7 +143,18 @@ int smcr_cdc_msg_send_validation(struct smc_connection *conn,
 	peer->token = htonl(local->token);
 	peer->prod_flags.failover_validation = 1;
 
+	/* We need to set pend->conn here to make sure smc_cdc_tx_handler()
+	 * can handle properly
+	 */
+	smc_cdc_add_pending_send(conn, pend);
+
+	atomic_inc(&conn->cdc_pend_tx_wr);
+	smp_mb__after_atomic(); /* Make sure cdc_pend_tx_wr added before post */
+
 	rc = smc_wr_tx_send(link, (struct smc_wr_tx_pend_priv *)pend);
+	if (unlikely(rc))
+		atomic_dec(&conn->cdc_pend_tx_wr);
+
 	return rc;
 }
 
@@ -193,31 +211,9 @@ int smc_cdc_get_slot_and_msg_send(struct smc_connection *conn)
 	return rc;
 }
 
-static bool smc_cdc_tx_filter(struct smc_wr_tx_pend_priv *tx_pend,
-			      unsigned long data)
+void smc_cdc_wait_pend_tx_wr(struct smc_connection *conn)
 {
-	struct smc_connection *conn = (struct smc_connection *)data;
-	struct smc_cdc_tx_pend *cdc_pend =
-		(struct smc_cdc_tx_pend *)tx_pend;
-
-	return cdc_pend->conn == conn;
-}
-
-static void smc_cdc_tx_dismisser(struct smc_wr_tx_pend_priv *tx_pend)
-{
-	struct smc_cdc_tx_pend *cdc_pend =
-		(struct smc_cdc_tx_pend *)tx_pend;
-
-	cdc_pend->conn = NULL;
-}
-
-void smc_cdc_tx_dismiss_slots(struct smc_connection *conn)
-{
-	struct smc_link *link = conn->lnk;
-
-	smc_wr_tx_dismiss_slots(link, SMC_CDC_MSG_TYPE,
-				smc_cdc_tx_filter, smc_cdc_tx_dismisser,
-				(unsigned long)conn);
+	wait_event(conn->cdc_pend_tx_wq, !atomic_read(&conn->cdc_pend_tx_wr));
 }
 
 /* Send a SMC-D CDC header.