ib_srp: bind fast IO failing to QP timeout

index dc49dc8..64644c5 100644
@@ -756,6 +756,29 @@ static void srp_reset_req(struct srp_target_port *target, struct srp_request *re
 	}
 }
 
+static void srp_fail_req(struct srp_target_port *target, struct srp_request *req)
+{
+	struct scsi_cmnd *scmnd = srp_claim_req(target, req, NULL);
+
+	if (scmnd) {
+		srp_free_req(target, req, scmnd, 0);
+		scmnd->result = DID_TRANSPORT_FAILFAST << 16;
+		scmnd->scsi_done(scmnd);
+	}
+}
+
+static void srp_fail_rport_io(struct srp_rport *rport)
+{
+	struct srp_target_port *target = rport->lld_data;
+	int i;
+
+	for (i = 0; i < SRP_CMD_SQ_SIZE; ++i) {
+		struct srp_request *req = &target->req_ring[i];
+		if (req->scmnd)
+			srp_fail_req(target, req);
+	}
+}
+
 static int srp_reconnect_target(struct srp_target_port *target)
 {
 	struct Scsi_Host *shost = target->scsi_host;

@@ -2700,6 +2723,7 @@ static void srp_remove_one(struct ib_device *device)
 
 static struct srp_function_template ib_srp_transport_functions = {
 	.rport_delete		 = srp_rport_delete,
+	.terminate_rport_io	 = srp_fail_rport_io,
 };
 
 static int __init srp_init_module(void)

index 64644c5..0607e5a 100644
@@ -750,6 +750,7 @@ static void srp_reset_req(struct srp_target_port *target, struct srp_request *re
 	struct scsi_cmnd *scmnd = srp_claim_req(target, req, NULL);
 
 	if (scmnd) {
+		scmnd->request->cmd_flags |= REQ_QUIET;
 		srp_free_req(target, req, scmnd, 0);
 		scmnd->result = DID_RESET << 16;
 		scmnd->scsi_done(scmnd);

@@ -761,6 +762,7 @@ static void srp_fail_req(struct srp_target_port *target, struct srp_request *req
 	struct scsi_cmnd *scmnd = srp_claim_req(target, req, NULL);
 
 	if (scmnd) {
+		scmnd->request->cmd_flags |= REQ_QUIET;
 		srp_free_req(target, req, scmnd, 0);
 		scmnd->result = DID_TRANSPORT_FAILFAST << 16;
 		scmnd->scsi_done(scmnd);

@@ -1526,6 +1528,7 @@ static int SRP_QUEUECOMMAND(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 	int len;
 
 	if (unlikely(target->transport_offline)) {
+		scmnd->request->cmd_flags |= REQ_QUIET;
 		scmnd->result = DID_NO_CONNECT << 16;
 		scmnd->scsi_done(scmnd);
 		return 0;

@@ -1934,6 +1937,7 @@ static int srp_abort(struct scsi_cmnd *scmnd)
 	else
 		ret = FAILED;
 
+	scmnd->request->cmd_flags |= REQ_QUIET;
 	srp_free_req(target, req, scmnd, 0);
 	scmnd->result = DID_ABORT << 16;
 	scmnd->scsi_done(scmnd);

index 239bb7a..be51bad 100644
@@ -312,10 +312,7 @@ static void rport_dev_loss_timedout(struct work_struct *work)
  */
 void srp_start_tl_fail_timers(struct srp_rport *rport, int elapsed_jiffies)
 {
-	if (rport->fast_io_fail_tmo >= 0)
-		queue_delayed_work(system_long_wq, &rport->fast_io_fail_work,
-				      max_t(long, 1UL * rport->fast_io_fail_tmo
-					    * HZ - elapsed_jiffies, 0));
+	queue_delayed_work(system_long_wq, &rport->fast_io_fail_work, 0);
 	queue_delayed_work(system_long_wq, &rport->dev_loss_work,
 		   max_t(long, 1UL * rport->dev_loss_tmo * HZ - elapsed_jiffies,
 			 0));

index 0607e5a..095c568 100644
@@ -2079,6 +2079,35 @@ static ssize_t show_allow_ext_sg(struct device *dev,
 	return sprintf(buf, "%s\n", target->allow_ext_sg ? "true" : "false");
 }
 
+static ssize_t show_qp_timeout_ms(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	struct srp_target_port *target = host_to_target(class_to_shost(dev));
+	struct ib_qp *qp = target->qp;
+	struct ib_qp_attr qp_attr;
+	struct ib_qp_init_attr qpi_attr;  /* useless, but required */
+	uint64_t T_tr_ms;
+
+	if (!qp || ib_query_qp(qp, &qp_attr, IB_QP_TIMEOUT, &qpi_attr) != 0)
+		return -EINVAL;
+	T_tr_ms = 4096 * (1ULL << qp_attr.timeout);
+	do_div(T_tr_ms, NSEC_PER_MSEC);
+	return sprintf(buf, "%llu\n", T_tr_ms);
+}
+
+static ssize_t show_qp_retries(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	struct srp_target_port *target = host_to_target(class_to_shost(dev));
+	struct ib_qp *qp = target->qp;
+	struct ib_qp_attr qp_attr;
+	struct ib_qp_init_attr qpi_attr;  /* useless, but required */
+
+	if (!qp || ib_query_qp(qp, &qp_attr, IB_QP_RETRY_CNT, &qpi_attr) != 0)
+		return -EINVAL;
+	return sprintf(buf, "%u\n", qp_attr.retry_cnt);
+}
+
 static DEVICE_ATTR(id_ext,	    S_IRUGO, show_id_ext,	   NULL);
 static DEVICE_ATTR(ioc_guid,	    S_IRUGO, show_ioc_guid,	   NULL);
 static DEVICE_ATTR(service_id,	    S_IRUGO, show_service_id,	   NULL);

@@ -2091,6 +2120,8 @@ static DEVICE_ATTR(local_ib_port,   S_IRUGO, show_local_ib_port,   NULL);
 static DEVICE_ATTR(local_ib_device, S_IRUGO, show_local_ib_device, NULL);
 static DEVICE_ATTR(cmd_sg_entries,  S_IRUGO, show_cmd_sg_entries,  NULL);
 static DEVICE_ATTR(allow_ext_sg,    S_IRUGO, show_allow_ext_sg,    NULL);
+static DEVICE_ATTR(qp_timeout_ms,   S_IRUGO, show_qp_timeout_ms,   NULL);
+static DEVICE_ATTR(qp_retries,      S_IRUGO, show_qp_retries,      NULL);
 
 static struct device_attribute *srp_host_attrs[] = {
 	&dev_attr_id_ext,

@@ -2105,6 +2136,8 @@ static struct device_attribute *srp_host_attrs[] = {
 	&dev_attr_local_ib_device,
 	&dev_attr_cmd_sg_entries,
 	&dev_attr_allow_ext_sg,
+	&dev_attr_qp_timeout_ms,
+	&dev_attr_qp_retries,
 	NULL
 };
 

index 095c568..c240fa7 100644
@@ -75,6 +75,8 @@ static unsigned int cmd_sg_entries;
 static unsigned int indirect_sg_entries;
 static bool allow_ext_sg;
 static int topspin_workarounds = 1;
+static unsigned int qp_retry_cnt;
+static bool qp_retry_set = false;
 
 module_param(srp_sg_tablesize, uint, 0444);
 MODULE_PARM_DESC(srp_sg_tablesize, "Deprecated name for cmd_sg_entries");

@@ -95,6 +97,10 @@ module_param(topspin_workarounds, int, 0444);
 MODULE_PARM_DESC(topspin_workarounds,
 		 "Enable workarounds for Topspin/Cisco SRP target bugs if != 0");
 
+module_param(qp_retry_cnt, uint, 0444);
+MODULE_PARM_DESC(qp_retry_cnt,
+		 "Number of qp retries, set this for fast IO failing (default is 7, max is 19)");
+
 static void srp_add_one(struct ib_device *device);
 static void srp_remove_one(struct ib_device *device);
 static void srp_recv_completion(struct ib_cq *cq, void *target_ptr);

@@ -1718,6 +1724,7 @@ static void srp_cm_rep_handler(struct ib_cm_id *cm_id,
 		goto error_free;
 
 	target->rq_tmo_jiffies = srp_compute_rq_tmo(qp_attr, attr_mask);
+	qp_attr->retry_cnt = qp_retry_cnt;
 
 	ret = ib_modify_qp(target->qp, qp_attr, attr_mask);
 	if (ret)

@@ -2183,6 +2190,7 @@ static int srp_add_target(struct srp_host *host, struct srp_target_port *target)
 	}
 
 	rport->lld_data = target;
+	rport->qp_retry_set = qp_retry_set;
 	target->rport = rport;
 
 	return 0;

@@ -2791,6 +2799,14 @@ static int __init srp_init_module(void)
 		indirect_sg_entries = cmd_sg_entries;
 	}
 
+	if (!qp_retry_cnt)
+		qp_retry_cnt = SRP_DEF_QP_RETRIES;
+	else
+		qp_retry_set = true;
+
+	if (qp_retry_cnt > SRP_MAX_QP_RETRIES)
+		qp_retry_cnt = SRP_MAX_QP_RETRIES;
+
 	ret = srp_transport_init();
 	if (ret) {
 		pr_err("SRP transport initialization failed\n");

index 1e8d538..ad43615 100644
@@ -57,6 +57,9 @@ enum {
 	SRP_MAX_LUN		= 512,
 	SRP_DEF_SG_TABLESIZE	= 12,
 
+	SRP_DEF_QP_RETRIES      = 7,
+	SRP_MAX_QP_RETRIES      = 19,
+
 	SRP_RQ_SHIFT    	= 6,
 	SRP_RQ_SIZE		= 1 << SRP_RQ_SHIFT,
 

index be51bad..347d231 100644
@@ -312,7 +312,8 @@ static void rport_dev_loss_timedout(struct work_struct *work)
  */
 void srp_start_tl_fail_timers(struct srp_rport *rport, int elapsed_jiffies)
 {
-	queue_delayed_work(system_long_wq, &rport->fast_io_fail_work, 0);
+	if (rport->qp_retry_set)
+		queue_delayed_work(system_long_wq, &rport->fast_io_fail_work, 0);
 	queue_delayed_work(system_long_wq, &rport->dev_loss_work,
 		   max_t(long, 1UL * rport->dev_loss_tmo * HZ - elapsed_jiffies,
 			 0));

index f74b715..aa9931e 100644
@@ -25,6 +25,7 @@ struct srp_rport {
 
 	void			*lld_data;	/* LLD private data */
 
+	bool			qp_retry_set;
 	int			fast_io_fail_tmo;
 	unsigned		dev_loss_tmo;
 	struct delayed_work	fast_io_fail_work;