[SCSI] libfc: Track rogue remote ports

Rogue ports are currently not tracked on any list. The only reference
to them is through any outstanding exchanges pending on the rogue ports.
If the module is removed while a retry is set on a rogue port
(say a Plogi retry for instance), this retry is not cancelled because there
is no reference to the rogue port in the discovery rports list. Thus the
local port can clean itself up, delete the exchange pool, and then the
rogue port timeout can fire and try to start up another exchange.

This patch tracks the rogue ports in a new list disc->rogue_rports. Creating
a new list instead of using the disc->rports list keeps remote port code
change to a minimum.

1)  Whenever a rogue port is created, it is immediately added to the
disc->rogue_rports list.

2) When the rogues port goes to ready, it is removed from the rogue list
and the real remote port is added to the disc->rports list

3) The removal of the rogue from the disc->rogue_rports list is done in
the context of the fc_rport_work() workQ thread in discovery callback.

4) Real rports are removed from the disc->rports list like before. Lookup
is done only in the real rports list. This avoids making large changes
to the remote port code.

5) In fc_disc_stop_rports, the rogues list is traversed in addition to the
real list to stop the rogue ports and issue logoffs on them. This way, rogue
ports get cleaned up when the local port goes away.

6) rogue remote ports are not removed from the list right away, but
removed late in fc_rport_work() context, multiple threads can find the same
remote port in the list and call rport_logoff(). Rport_logoff() only
continues with the logoff if port is not in NONE state, thus preventing
multiple logoffs and multiple list deletions.

7) Since the rport is removed from the disc list at a later stage
(in the disc callback), incoming frames can find the rport even if
rport_logoff() has been called on the rport. When rport_logoff() is called,
the rport state is set to NONE, and we are trying to cancel all exchanges
and retries on that port. While in this state, if an incoming
Plogi/Prli/Logo or other frames match the rport, we should not reply
because the rport is in the NONE state. Just drop the frame, since the
rport will be deleted soon in the disc callback (fc_rport_work)

8)  In fc_disc_single(), remove rport lookup and call to fc_disc_del_target.
fc_disc_single() is called from recv_rscn_req() where rport lookup
and rport_logoff is already done.

Signed-off-by: Abhijeet Joglekar <abjoglek@cisco.com>
Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>

1 files changed, 25 insertions, 11 deletions

diff --git a/drivers/scsi/libfc/fc_disc.c b/drivers/scsi/libfc/fc_disc.c
index 44806307f831..4c880656990b 100644
--- a/drivers/scsi/libfc/fc_disc.c
+++ b/drivers/scsi/libfc/fc_disc.c
@@ -113,6 +113,11 @@ void fc_disc_stop_rports(struct fc_disc *disc)
 		lport->tt.rport_logoff(rport);
 	}
 
+	list_for_each_entry_safe(rdata, next, &disc->rogue_rports, peers) {
+		rport = PRIV_TO_RPORT(rdata);
+		lport->tt.rport_logoff(rport);
+	}
+
 	mutex_unlock(&disc->disc_mutex);
 }
 
@@ -131,23 +136,32 @@ static void fc_disc_rport_callback(struct fc_lport *lport,
 {
 	struct fc_rport_libfc_priv *rdata = rport->dd_data;
 	struct fc_disc *disc = &lport->disc;
-	int found = 0;
 
 	FC_DEBUG_DISC("Received a %d event for port (%6x)\n", event,
 		      rport->port_id);
 
-	if (event == RPORT_EV_CREATED) {
+	switch (event) {
+	case RPORT_EV_CREATED:
 		if (disc) {
-			found = 1;
 			mutex_lock(&disc->disc_mutex);
 			list_add_tail(&rdata->peers, &disc->rports);
 			mutex_unlock(&disc->disc_mutex);
 		}
+		break;
+	case RPORT_EV_LOGO:
+	case RPORT_EV_FAILED:
+	case RPORT_EV_STOP:
+		mutex_lock(&disc->disc_mutex);
+		mutex_lock(&rdata->rp_mutex);
+		if (rdata->trans_state == FC_PORTSTATE_ROGUE)
+			list_del(&rdata->peers);
+		mutex_unlock(&rdata->rp_mutex);
+		mutex_unlock(&disc->disc_mutex);
+		break;
+	default:
+		break;
 	}
 
-	if (!found)
-		FC_DEBUG_DISC("The rport (%6x) is not maintained "
-			      "by the discovery layer\n", rport->port_id);
 }
 
 /**
@@ -439,6 +453,7 @@ static int fc_disc_new_target(struct fc_disc *disc,
 			rdata = rport->dd_data;
 			rdata->ops = &fc_disc_rport_ops;
 			rdata->rp_state = RPORT_ST_INIT;
+			list_add_tail(&rdata->peers, &disc->rogue_rports);
 			lport->tt.rport_login(rport);
 		}
 	}
@@ -630,6 +645,8 @@ static int fc_disc_gpn_ft_parse(struct fc_disc *disc, void *buf, size_t len)
 				rdata = rport->dd_data;
 				rdata->ops = &fc_disc_rport_ops;
 				rdata->local_port = lport;
+				list_add_tail(&rdata->peers,
+					      &disc->rogue_rports);
 				lport->tt.rport_login(rport);
 			} else
 				FC_DBG("Failed to allocate memory for "
@@ -769,7 +786,6 @@ static void fc_disc_gpn_ft_resp(struct fc_seq *sp, struct fc_frame *fp,
 static void fc_disc_single(struct fc_disc *disc, struct fc_disc_port *dp)
 {
 	struct fc_lport *lport;
-	struct fc_rport *rport;
 	struct fc_rport *new_rport;
 	struct fc_rport_libfc_priv *rdata;
 
@@ -778,15 +794,12 @@ static void fc_disc_single(struct fc_disc *disc, struct fc_disc_port *dp)
 	if (dp->ids.port_id == fc_host_port_id(lport->host))
 		goto out;
 
-	rport = lport->tt.rport_lookup(lport, dp->ids.port_id);
-	if (rport)
-		fc_disc_del_target(disc, rport);
-
 	new_rport = lport->tt.rport_create(dp);
 	if (new_rport) {
 		rdata = new_rport->dd_data;
 		rdata->ops = &fc_disc_rport_ops;
 		kfree(dp);
+		list_add_tail(&rdata->peers, &disc->rogue_rports);
 		lport->tt.rport_login(new_rport);
 	}
 	return;
@@ -848,6 +861,7 @@ int fc_disc_init(struct fc_lport *lport)
 	INIT_DELAYED_WORK(&disc->disc_work, fc_disc_timeout);
 	mutex_init(&disc->disc_mutex);
 	INIT_LIST_HEAD(&disc->rports);
+	INIT_LIST_HEAD(&disc->rogue_rports);
 
 	disc->lport = lport;
 	disc->delay = FC_DISC_DELAY;