[patch 00/15] zfcp fixes for 2.6.30-rc2

[patch 00/15] zfcp fixes for 2.6.30-rc2

[Christof Schmitt]

James,

here is the series of zfcp fixes for the current development cycle.
The patches are based on top of the current linux-2.6 kernel
(2.6.30-rc2).

--
Christof Schmitt

[patch 01/15] zfcp: Avoid referencing freed memory in req send

[Christof Schmitt]

[-- Attachment #1: 700-zfcp-reference-after-free.diff --]
[-- Type: text/plain, Size: 1132 bytes --]

From: Martin Petermann <martin@linux.vnet.ibm.com>

Avoid referencing a fsf request after sending it in fcp_fsf_req_send,
it might have already completed and deallocated.

Signed-off-by: Martin Petermann <martin@linux.vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fsf.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

--- a/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:22.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:25.000000000 +0200
@@ -766,8 +766,9 @@ static struct zfcp_fsf_req *zfcp_fsf_req
 static int zfcp_fsf_req_send(struct zfcp_fsf_req *req)
 {
 	struct zfcp_adapter *adapter = req->adapter;
-	unsigned long flags;
-	int idx;
+	unsigned long	     flags;
+	int		     idx;
+	int		     with_qtcb = (req->qtcb != NULL);
 
 	/* put allocated FSF request into hash table */
 	spin_lock_irqsave(&adapter->req_list_lock, flags);
@@ -789,7 +790,7 @@ static int zfcp_fsf_req_send(struct zfcp
 	}
 
 	/* Don't increase for unsolicited status */
-	if (req->qtcb)
+	if (with_qtcb)
 		adapter->fsf_req_seq_no++;
 	adapter->req_no++;
 

[patch 02/15] zfcp: Enable auto-port discovery for NPIV.

[Christof Schmitt]

[-- Attachment #1: 701-zfcp-port-discovery.diff --]
[-- Type: text/plain, Size: 1005 bytes --]

From: Swen Schillig <swen@vnet.ibm.com>

Since we're setting the host port type now to FC_PORTTYPE_NPIV
for adapters running in NPIV mode we should allow this port type
for auto-port scanning as well.

Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fc.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:22.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:31.000000000 +0200
@@ -637,7 +637,8 @@ int zfcp_scan_ports(struct zfcp_adapter 
 	max_entries = chain ? ZFCP_GPN_FT_MAX_ENTRIES : ZFCP_GPN_FT_ENTRIES;
 	max_bytes = chain ? ZFCP_GPN_FT_MAX_SIZE : ZFCP_CT_SIZE_ONE_PAGE;
 
-	if (fc_host_port_type(adapter->scsi_host) != FC_PORTTYPE_NPORT)
+	if (fc_host_port_type(adapter->scsi_host) != FC_PORTTYPE_NPORT &&
+	    fc_host_port_type(adapter->scsi_host) != FC_PORTTYPE_NPIV)
 		return 0;
 
 	ret = zfcp_wka_port_get(&adapter->nsp);

[patch 03/15] zfcp: Dont call zfcp_fsf_req_free on NULL pointer

[Christof Schmitt]

[-- Attachment #1: 702-zfcp-fsf-req-free.diff --]
[-- Type: text/plain, Size: 2293 bytes --]

From: Christof Schmitt <christof.schmitt@de.ibm.com>

Fix problem that zfcp_fsf_exchange_config_data_sync and
zfcp_fsf_exchange_config_data_sync could try to call zfcp_fsf_req_free
with a NULL pointer.

Reviewed-by: Martin Petermann <martin@linux.vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fsf.c |   18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

--- a/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:25.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:33.000000000 +0200
@@ -1254,13 +1254,13 @@ int zfcp_fsf_exchange_config_data_sync(s
 
 	spin_lock_bh(&adapter->req_q_lock);
 	if (zfcp_fsf_req_sbal_get(adapter))
-		goto out;
+		goto out_unlock;
 
 	req = zfcp_fsf_req_create(adapter, FSF_QTCB_EXCHANGE_CONFIG_DATA,
 				  0, NULL);
 	if (IS_ERR(req)) {
 		retval = PTR_ERR(req);
-		goto out;
+		goto out_unlock;
 	}
 
 	sbale = zfcp_qdio_sbale_req(req);
@@ -1279,14 +1279,16 @@ int zfcp_fsf_exchange_config_data_sync(s
 
 	zfcp_fsf_start_timer(req, ZFCP_FSF_REQUEST_TIMEOUT);
 	retval = zfcp_fsf_req_send(req);
-out:
 	spin_unlock_bh(&adapter->req_q_lock);
 	if (!retval)
 		wait_event(req->completion_wq,
 			   req->status & ZFCP_STATUS_FSFREQ_COMPLETED);
 
 	zfcp_fsf_req_free(req);
+	return retval;
 
+out_unlock:
+	spin_unlock_bh(&adapter->req_q_lock);
 	return retval;
 }
 
@@ -1353,13 +1355,13 @@ int zfcp_fsf_exchange_port_data_sync(str
 
 	spin_lock_bh(&adapter->req_q_lock);
 	if (zfcp_fsf_req_sbal_get(adapter))
-		goto out;
+		goto out_unlock;
 
 	req = zfcp_fsf_req_create(adapter, FSF_QTCB_EXCHANGE_PORT_DATA, 0,
 				  NULL);
 	if (IS_ERR(req)) {
 		retval = PTR_ERR(req);
-		goto out;
+		goto out_unlock;
 	}
 
 	if (data)
@@ -1372,14 +1374,18 @@ int zfcp_fsf_exchange_port_data_sync(str
 	req->handler = zfcp_fsf_exchange_port_data_handler;
 	zfcp_fsf_start_timer(req, ZFCP_FSF_REQUEST_TIMEOUT);
 	retval = zfcp_fsf_req_send(req);
-out:
 	spin_unlock_bh(&adapter->req_q_lock);
+
 	if (!retval)
 		wait_event(req->completion_wq,
 			   req->status & ZFCP_STATUS_FSFREQ_COMPLETED);
 	zfcp_fsf_req_free(req);
 
 	return retval;
+
+out_unlock:
+	spin_unlock_bh(&adapter->req_q_lock);
+	return retval;
 }
 
 static void zfcp_fsf_open_port_handler(struct zfcp_fsf_req *req)

[patch 04/15] zfcp: Dont block zfcp_wq with scan

[Christof Schmitt]

[-- Attachment #1: scsi-work.diff --]
[-- Type: text/plain, Size: 5590 bytes --]

From: Swen Schillig <swen@vnet.ibm.com>

When running the scsi_scan from the zfcp workqueue and the target
device does not respond, the zfcp workqueue can block until the
scsi_scan hits a timeout. Move the work to the scsi host workqueue,
since this one is also used for the scan from the SCSI midlayer.

Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---
 drivers/s390/scsi/zfcp_aux.c  |    5 +---
 drivers/s390/scsi/zfcp_def.h  |    2 -
 drivers/s390/scsi/zfcp_erp.c  |   50 +++---------------------------------------
 drivers/s390/scsi/zfcp_ext.h  |    1 
 drivers/s390/scsi/zfcp_scsi.c |   17 ++++++++++++++
 5 files changed, 25 insertions(+), 50 deletions(-)

--- a/drivers/s390/scsi/zfcp_aux.c	2009-04-17 15:03:21.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_aux.c	2009-04-17 15:03:35.000000000 +0200
@@ -97,9 +97,7 @@ static void __init zfcp_init_device_conf
 	ccw_device_set_online(adapter->ccw_device);
 
 	zfcp_erp_wait(adapter);
-	wait_event(adapter->erp_done_wqh,
-		   !(atomic_read(&unit->status) &
-				ZFCP_STATUS_UNIT_SCSI_WORK_PENDING));
+	flush_work(&unit->scsi_work);
 
 	down(&zfcp_data.config_sema);
 	zfcp_unit_put(unit);
@@ -279,6 +277,7 @@ struct zfcp_unit *zfcp_unit_enqueue(stru
 
 	atomic_set(&unit->refcount, 0);
 	init_waitqueue_head(&unit->remove_wq);
+	INIT_WORK(&unit->scsi_work, zfcp_scsi_scan);
 
 	unit->port = port;
 	unit->fcp_lun = fcp_lun;
--- a/drivers/s390/scsi/zfcp_def.h	2009-04-17 15:03:21.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_def.h	2009-04-17 15:03:35.000000000 +0200
@@ -255,7 +255,6 @@ enum zfcp_wka_status {
 /* logical unit status */
 #define ZFCP_STATUS_UNIT_SHARED			0x00000004
 #define ZFCP_STATUS_UNIT_READONLY		0x00000008
-#define ZFCP_STATUS_UNIT_SCSI_WORK_PENDING	0x00000020
 
 /* FSF request status (this does not have a common part) */
 #define ZFCP_STATUS_FSFREQ_TASK_MANAGEMENT	0x00000002
@@ -530,6 +529,7 @@ struct zfcp_unit {
 	struct zfcp_erp_action erp_action;     /* pending error recovery */
         atomic_t               erp_counter;
 	struct zfcp_latencies	latencies;
+	struct work_struct	scsi_work;
 };
 
 /* FSF request */
--- a/drivers/s390/scsi/zfcp_erp.c	2009-04-17 15:03:21.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_erp.c	2009-04-17 15:03:35.000000000 +0200
@@ -1176,48 +1176,6 @@ static void zfcp_erp_action_dequeue(stru
 	}
 }
 
-struct zfcp_erp_add_work {
-	struct zfcp_unit  *unit;
-	struct work_struct work;
-};
-
-static void zfcp_erp_scsi_scan(struct work_struct *work)
-{
-	struct zfcp_erp_add_work *p =
-		container_of(work, struct zfcp_erp_add_work, work);
-	struct zfcp_unit *unit = p->unit;
-	struct fc_rport *rport = unit->port->rport;
-
-	if (rport && rport->port_state == FC_PORTSTATE_ONLINE)
-		scsi_scan_target(&rport->dev, 0, rport->scsi_target_id,
-			 scsilun_to_int((struct scsi_lun *)&unit->fcp_lun), 0);
-	atomic_clear_mask(ZFCP_STATUS_UNIT_SCSI_WORK_PENDING, &unit->status);
-	zfcp_unit_put(unit);
-	wake_up(&unit->port->adapter->erp_done_wqh);
-	kfree(p);
-}
-
-static void zfcp_erp_schedule_work(struct zfcp_unit *unit)
-{
-	struct zfcp_erp_add_work *p;
-
-	p = kzalloc(sizeof(*p), GFP_KERNEL);
-	if (!p) {
-		dev_err(&unit->port->adapter->ccw_device->dev,
-			"Registering unit 0x%016Lx on port 0x%016Lx failed\n",
-			(unsigned long long)unit->fcp_lun,
-			(unsigned long long)unit->port->wwpn);
-		return;
-	}
-
-	zfcp_unit_get(unit);
-	atomic_set_mask(ZFCP_STATUS_UNIT_SCSI_WORK_PENDING, &unit->status);
-	INIT_WORK(&p->work, zfcp_erp_scsi_scan);
-	p->unit = unit;
-	if (!queue_work(zfcp_data.work_queue, &p->work))
-		zfcp_unit_put(unit);
-}
-
 static void zfcp_erp_action_cleanup(struct zfcp_erp_action *act, int result)
 {
 	struct zfcp_adapter *adapter = act->adapter;
@@ -1226,11 +1184,11 @@ static void zfcp_erp_action_cleanup(stru
 
 	switch (act->action) {
 	case ZFCP_ERP_ACTION_REOPEN_UNIT:
-		flush_work(&port->rport_work);
 		if ((result == ZFCP_ERP_SUCCEEDED) && !unit->device) {
-			if (!(atomic_read(&unit->status) &
-			      ZFCP_STATUS_UNIT_SCSI_WORK_PENDING))
-				zfcp_erp_schedule_work(unit);
+			zfcp_unit_get(unit);
+			if (scsi_queue_work(unit->port->adapter->scsi_host,
+					    &unit->scsi_work) <= 0)
+				zfcp_unit_put(unit);
 		}
 		zfcp_unit_put(unit);
 		break;
--- a/drivers/s390/scsi/zfcp_ext.h	2009-04-17 15:03:21.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_ext.h	2009-04-17 15:03:35.000000000 +0200
@@ -158,6 +158,7 @@ extern void zfcp_scsi_rport_work(struct 
 extern void zfcp_scsi_schedule_rport_register(struct zfcp_port *);
 extern void zfcp_scsi_schedule_rport_block(struct zfcp_port *);
 extern void zfcp_scsi_schedule_rports_block(struct zfcp_adapter *);
+extern void zfcp_scsi_scan(struct work_struct *);
 
 /* zfcp_sysfs.c */
 extern struct attribute_group zfcp_sysfs_unit_attrs;
--- a/drivers/s390/scsi/zfcp_scsi.c	2009-04-17 15:03:21.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_scsi.c	2009-04-17 15:03:35.000000000 +0200
@@ -583,6 +583,23 @@ void zfcp_scsi_rport_work(struct work_st
 }
 
 
+void zfcp_scsi_scan(struct work_struct *work)
+{
+	struct zfcp_unit *unit = container_of(work, struct zfcp_unit,
+					      scsi_work);
+	struct fc_rport *rport;
+
+	flush_work(&unit->port->rport_work);
+	rport = unit->port->rport;
+
+	if (rport && rport->port_state == FC_PORTSTATE_ONLINE)
+		scsi_scan_target(&rport->dev, 0, rport->scsi_target_id,
+				 scsilun_to_int((struct scsi_lun *)
+						&unit->fcp_lun), 0);
+
+	zfcp_unit_put(unit);
+}
+
 struct fc_function_template zfcp_transport_functions = {
 	.show_starget_port_id = 1,
 	.show_starget_port_name = 1,

[patch 05/15] zfcp: Set WKA-port to offline on adapter deactivation

[Christof Schmitt]

[-- Attachment #1: 704.diff --]
[-- Type: text/plain, Size: 4094 bytes --]

From: Swen Schillig <swen@vnet.ibm.com>

The nameserver port might be in state online when the adapter is
offlined. On adapter reactivation the nameserver port is not
re-opened due to the PORT_ONLINE status. This results in an
unsuccessful recovery. In forcing the nameserver port status
to offline on all adapter offline events this issue is prevented.

Waiting for the reference count to drop to zero in
zfcp_wka_port_offline is not required, so remove it.

Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---
 drivers/s390/scsi/zfcp_aux.c |    3 ++-
 drivers/s390/scsi/zfcp_ccw.c |    1 -
 drivers/s390/scsi/zfcp_erp.c |    1 +
 drivers/s390/scsi/zfcp_ext.h |    1 +
 drivers/s390/scsi/zfcp_fc.c  |   15 ++++++++-------
 5 files changed, 12 insertions(+), 9 deletions(-)

--- a/drivers/s390/scsi/zfcp_aux.c	2009-04-17 15:03:35.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_aux.c	2009-04-17 15:03:37.000000000 +0200
@@ -524,6 +524,8 @@ int zfcp_adapter_enqueue(struct ccw_devi
 
 	atomic_clear_mask(ZFCP_STATUS_COMMON_REMOVE, &adapter->status);
 
+	zfcp_fc_nameserver_init(adapter);
+
 	if (!zfcp_adapter_scsi_register(adapter))
 		return 0;
 
@@ -552,7 +554,6 @@ void zfcp_adapter_dequeue(struct zfcp_ad
 
 	cancel_work_sync(&adapter->scan_work);
 	cancel_work_sync(&adapter->stat_work);
-	cancel_delayed_work_sync(&adapter->nsp.work);
 	zfcp_adapter_scsi_unregister(adapter);
 	sysfs_remove_group(&adapter->ccw_device->dev.kobj,
 			   &zfcp_sysfs_adapter_attrs);
--- a/drivers/s390/scsi/zfcp_ccw.c	2009-04-17 15:03:20.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_ccw.c	2009-04-17 15:03:37.000000000 +0200
@@ -108,7 +108,6 @@ static int zfcp_ccw_set_online(struct cc
 	/* initialize request counter */
 	BUG_ON(!zfcp_reqlist_isempty(adapter));
 	adapter->req_no = 0;
-	zfcp_fc_nameserver_init(adapter);
 
 	zfcp_erp_modify_adapter_status(adapter, "ccsonl1", NULL,
 				       ZFCP_STATUS_COMMON_RUNNING, ZFCP_SET);
--- a/drivers/s390/scsi/zfcp_erp.c	2009-04-17 15:03:35.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_erp.c	2009-04-17 15:03:37.000000000 +0200
@@ -719,6 +719,7 @@ static void zfcp_erp_adapter_strategy_cl
 	zfcp_qdio_close(adapter);
 	zfcp_fsf_req_dismiss_all(adapter);
 	adapter->fsf_req_seq_no = 0;
+	zfcp_fc_wka_port_force_offline(&adapter->nsp);
 	/* all ports and units are closed */
 	zfcp_erp_modify_adapter_status(adapter, "erascl1", NULL,
 				       ZFCP_STATUS_COMMON_OPEN, ZFCP_CLEAR);
--- a/drivers/s390/scsi/zfcp_ext.h	2009-04-17 15:03:35.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_ext.h	2009-04-17 15:03:37.000000000 +0200
@@ -106,6 +106,7 @@ extern void zfcp_fc_plogi_evaluate(struc
 extern void zfcp_test_link(struct zfcp_port *);
 extern void zfcp_fc_link_test_work(struct work_struct *);
 extern void zfcp_fc_nameserver_init(struct zfcp_adapter *);
+extern void zfcp_fc_wka_port_force_offline(struct zfcp_wka_port *);
 
 /* zfcp_fsf.c */
 extern int zfcp_fsf_open_port(struct zfcp_erp_action *);
--- a/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:31.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:37.000000000 +0200
@@ -98,13 +98,6 @@ static void zfcp_wka_port_offline(struct
 	struct zfcp_wka_port *wka_port =
 			container_of(dw, struct zfcp_wka_port, work);
 
-	/* Don't wait forvever. If the wka_port is too busy take it offline
-	   through a new call later */
-	if (!wait_event_timeout(wka_port->completion_wq,
-				atomic_read(&wka_port->refcount) == 0,
-				HZ >> 1))
-		return;
-
 	mutex_lock(&wka_port->mutex);
 	if ((atomic_read(&wka_port->refcount) != 0) ||
 	    (wka_port->status != ZFCP_WKA_PORT_ONLINE))
@@ -142,6 +135,14 @@ void zfcp_fc_nameserver_init(struct zfcp
 	INIT_DELAYED_WORK(&wka_port->work, zfcp_wka_port_offline);
 }
 
+void zfcp_fc_wka_port_force_offline(struct zfcp_wka_port *wka)
+{
+	cancel_delayed_work_sync(&wka->work);
+	mutex_lock(&wka->mutex);
+	wka->status = ZFCP_WKA_PORT_OFFLINE;
+	mutex_unlock(&wka->mutex);
+}
+
 static void _zfcp_fc_incoming_rscn(struct zfcp_fsf_req *fsf_req, u32 range,
 				   struct fcp_rscn_element *elem)
 {

[patch 06/15] zfcp: avoid false ERP complete due to sema race

[Christof Schmitt]

[-- Attachment #1: 705-zfcp-false-complete.diff --]
[-- Type: text/plain, Size: 1845 bytes --]

From: Swen Schillig <swen@vnet.ibm.com>

The ERP thread is performing a task before it is executing the
corresponding down on the semaphore. The response handler of the
just started exchange config should wait for the completion by
performing a down on this semaphore. Since this semaphore is still
positive from the ERP enqueue the handler won't wait and therefore
the exchange config will always fail leaving the adapter in error.
The problem can be solved by performing the down on the semaphore 
before starting an ERP task. This is the logically correct order. 
Only walk the ERP loop if there is a task to perform.

Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_erp.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

--- a/drivers/s390/scsi/zfcp_erp.c	2009-04-17 15:03:37.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_erp.c	2009-04-17 15:03:39.000000000 +0200
@@ -1311,6 +1311,11 @@ static int zfcp_erp_thread(void *data)
 
 	while (!(atomic_read(&adapter->status) &
 		 ZFCP_STATUS_ADAPTER_ERP_THREAD_KILL)) {
+
+		zfcp_rec_dbf_event_thread_lock("erthrd1", adapter);
+		ignore = down_interruptible(&adapter->erp_ready_sem);
+		zfcp_rec_dbf_event_thread_lock("erthrd2", adapter);
+
 		write_lock_irqsave(&adapter->erp_lock, flags);
 		next = adapter->erp_ready_head.next;
 		write_unlock_irqrestore(&adapter->erp_lock, flags);
@@ -1322,10 +1327,6 @@ static int zfcp_erp_thread(void *data)
 			if (zfcp_erp_strategy(act) != ZFCP_ERP_DISMISSED)
 				zfcp_erp_wakeup(adapter);
 		}
-
-		zfcp_rec_dbf_event_thread_lock("erthrd1", adapter);
-		ignore = down_interruptible(&adapter->erp_ready_sem);
-		zfcp_rec_dbf_event_thread_lock("erthrd2", adapter);
 	}
 
 	atomic_clear_mask(ZFCP_STATUS_ADAPTER_ERP_THREAD_UP, &adapter->status);

[patch 07/15] zfcp: no port recovery after storage side error inject

[Christof Schmitt]

[-- Attachment #1: 707-zfcp-port-recovery.diff --]
[-- Type: text/plain, Size: 1045 bytes --]

From: Swen Schillig <swen@vnet.ibm.com>

The remote port remains in error state even if the connection
is re-established. A wrong precondition check was performed on
the port status leading to a cancellation of the port reopen.
Remove the pre-req check because it's not required and better
handled within the ERP.

Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fc.c |    5 -----
 1 file changed, 5 deletions(-)

--- a/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:37.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:40.000000000 +0200
@@ -432,11 +432,6 @@ void zfcp_fc_link_test_work(struct work_
 		container_of(work, struct zfcp_port, test_link_work);
 	int retval;
 
-	if (!(atomic_read(&port->status) & ZFCP_STATUS_COMMON_UNBLOCKED)) {
-		zfcp_port_put(port);
-		return; /* port erp is running and will update rport status */
-	}
-
 	zfcp_port_get(port);
 	port->rport_task = RPORT_DEL;
 	zfcp_scsi_rport_work(&port->rport_work);

[patch 08/15] zfcp: remove unit will fail if add unit is not finished

[Christof Schmitt]

[-- Attachment #1: 708-zfcp-remove-unit.diff --]
[-- Type: text/plain, Size: 1379 bytes --]

From: Martin Petermann <martin@linux.vnet.ibm.com>

On some hardware it can take some time to add a unit. If
some remove this unit during this process the remove will
fail.

Signed-off-by: Martin Petermann <martin@linux.vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_sysfs.c |   21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

--- a/drivers/s390/scsi/zfcp_sysfs.c	2009-04-17 15:03:20.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_sysfs.c	2009-04-17 15:03:42.000000000 +0200
@@ -254,12 +254,21 @@ static ssize_t zfcp_sysfs_unit_remove_st
 
 	write_lock_irq(&zfcp_data.config_lock);
 	unit = zfcp_get_unit_by_lun(port, fcp_lun);
-	if (unit && (atomic_read(&unit->refcount) == 0)) {
-		zfcp_unit_get(unit);
-		atomic_set_mask(ZFCP_STATUS_COMMON_REMOVE, &unit->status);
-		list_move(&unit->list, &unit_remove_lh);
-	} else
-		unit = NULL;
+	if (unit) {
+		write_unlock_irq(&zfcp_data.config_lock);
+		/* wait for possible timeout during SCSI probe */
+		flush_work(&unit->scsi_work);
+		write_lock_irq(&zfcp_data.config_lock);
+
+		if (atomic_read(&unit->refcount) == 0) {
+			zfcp_unit_get(unit);
+			atomic_set_mask(ZFCP_STATUS_COMMON_REMOVE,
+					&unit->status);
+			list_move(&unit->list, &unit_remove_lh);
+		} else {
+			unit = NULL;
+		}
+	}
 
 	write_unlock_irq(&zfcp_data.config_lock);
 

[patch 09/15] zfcp: Let actcli handle control file errors

[Christof Schmitt]

[-- Attachment #1: 709-zfcp-actcli-errors.diff --]
[-- Type: text/plain, Size: 793 bytes --]

From: Christof Schmitt <christof.schmitt@de.ibm.com>

Error codes specific to the control file requests are evaluated by the
actcli tool, so don't report -ENXIO for those. Generic problems are
still checked for outside the command specific handler.

Reviewed-by: Martin Petermann <martin@linux.vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fsf.c |    2 --
 1 file changed, 2 deletions(-)

--- a/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:33.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:44.000000000 +0200
@@ -2479,8 +2479,6 @@ out:
 
 static void zfcp_fsf_control_file_handler(struct zfcp_fsf_req *req)
 {
-	if (req->qtcb->header.fsf_status != FSF_GOOD)
-		req->status |= ZFCP_STATUS_FSFREQ_ERROR;
 }
 
 /**

[patch 10/15] zfcp: no port recovery after ADISC request timeout

[Christof Schmitt]

[-- Attachment #1: 710-zfcp-adisc-timeout.diff --]
[-- Type: text/plain, Size: 937 bytes --]

From: Swen Schillig <swen@vnet.ibm.com>

A remote port remains in error state even if we receive a RSCN
stating that the connection is re-established. The port recovery
is not started due to a flag which is not reset.
The solution is to clear the flag in question before we trigger a ERP.

Signed-off-by: Swen Schillig <swen@vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fc.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:40.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:46.000000000 +0200
@@ -373,7 +373,8 @@ static void zfcp_fc_adisc_handler(unsign
 
 	if (adisc->els.status) {
 		/* request rejected or timed out */
-		zfcp_erp_port_forced_reopen(port, 0, "fcadh_1", NULL);
+		zfcp_erp_port_forced_reopen(port, ZFCP_STATUS_COMMON_ERP_FAILED,
+					    "fcadh_1", NULL);
 		goto out;
 	}
 

[patch 11/15] zfcp: Fix abort handler for completions in progress

[Christof Schmitt]

[-- Attachment #1: 711-zfcp-abort-handler.diff --]
[-- Type: text/plain, Size: 959 bytes --]

From: Christof Schmitt <christof.schmitt@de.ibm.com>

When the abort handler cannot find a pending FSF request, the request
completion could just be running. This means we cannot return SUCCESS,
since this would lead to call to scsi_done after exiting the SCSI
error handler which is not allowed.

Reviewed-by: Martin Petermann <martin@linux.vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_scsi.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/s390/scsi/zfcp_scsi.c	2009-04-17 15:03:35.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_scsi.c	2009-04-17 15:03:49.000000000 +0200
@@ -171,7 +171,7 @@ static int zfcp_scsi_eh_abort_handler(st
 		write_unlock_irqrestore(&adapter->abort_lock, flags);
 		zfcp_scsi_dbf_event_abort("lte1", adapter, scpnt, NULL,
 					  old_req_id);
-		return SUCCESS;
+		return FAILED; /* completion could be in progress */
 	}
 	old_req->data = NULL;
 

[patch 12/15] zfcp: revert previous patch for sbal counting

[Christof Schmitt]

[-- Attachment #1: 712-zfcp-sbal-counting.diff --]
[-- Type: text/plain, Size: 1764 bytes --]

From: Martin Petermann <martin.petermann@de.ibm.com>

The current sbal counting can be wrong if a fsf request is
waiting for free sbals and at the same time qdio request queue
is shutdown and re-opened. Revering a previous patch fixes this
issue.

Signed-off-by: Martin Petermann <martin.petermann@de.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fsf.c |   28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

--- a/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:44.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:52.000000000 +0200
@@ -645,30 +645,30 @@ static void zfcp_fsf_exchange_port_data_
 	}
 }
 
-static int zfcp_fsf_req_sbal_get(struct zfcp_adapter *adapter)
-	__releases(&adapter->req_q_lock)
-	__acquires(&adapter->req_q_lock)
+static int zfcp_fsf_sbal_check(struct zfcp_adapter *adapter)
 {
 	struct zfcp_qdio_queue *req_q = &adapter->req_q;
-	long ret;
 
-	if (atomic_read(&req_q->count) <= -REQUEST_LIST_SIZE)
-		return -EIO;
-	if (atomic_read(&req_q->count) > 0)
-		return 0;
+	spin_lock_bh(&adapter->req_q_lock);
+	if (atomic_read(&req_q->count))
+		return 1;
+	spin_unlock_bh(&adapter->req_q_lock);
+	return 0;
+}
+
+static int zfcp_fsf_req_sbal_get(struct zfcp_adapter *adapter)
+{
+	long ret;
 
-	atomic_dec(&req_q->count);
 	spin_unlock_bh(&adapter->req_q_lock);
 	ret = wait_event_interruptible_timeout(adapter->request_wq,
-					atomic_read(&req_q->count) >= 0,
-					5 * HZ);
-	spin_lock_bh(&adapter->req_q_lock);
-	atomic_inc(&req_q->count);
-
+			       zfcp_fsf_sbal_check(adapter), 5 * HZ);
 	if (ret > 0)
 		return 0;
 	if (!ret)
 		atomic_inc(&adapter->qdio_outb_full);
+
+	spin_lock_bh(&adapter->req_q_lock);
 	return -EIO;
 }
 

[patch 13/15] zfcp: Fix port reference counting

[Christof Schmitt]

[-- Attachment #1: 713-zfcp-port-refcounting.diff --]
[-- Type: text/plain, Size: 1349 bytes --]

From: Martin Petermann <martin.petermann@de.ibm.com>

If this problem appears zfcp ports cannot be de-queued since it is 
checked for a zero refcount. The port reference counting is wrong for
existing zfcp ports when e.g. an adapter gets on-line again. During 
port scanning the reference counting for existing ports should not be
changed.

Signed-off-by: Martin Petermann <martin.petermann@de.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---

 drivers/s390/scsi/zfcp_fc.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

--- a/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:46.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fc.c	2009-04-17 15:03:54.000000000 +0200
@@ -539,6 +539,9 @@ static void zfcp_validate_port(struct zf
 {
 	struct zfcp_adapter *adapter = port->adapter;
 
+	if (!(atomic_read(&port->status) & ZFCP_STATUS_COMMON_NOESC))
+		return;
+
 	atomic_clear_mask(ZFCP_STATUS_COMMON_NOESC, &port->status);
 
 	if ((port->supported_classes != 0) ||
@@ -599,10 +602,8 @@ static int zfcp_scan_eval_gpn_ft(struct 
 		if (acc->wwpn == fc_host_port_name(adapter->scsi_host))
 			continue;
 		port = zfcp_get_port_by_wwpn(adapter, acc->wwpn);
-		if (port) {
-			zfcp_port_get(port);
+		if (port)
 			continue;
-		}
 
 		port = zfcp_port_enqueue(adapter, acc->wwpn,
 					 ZFCP_STATUS_COMMON_NOESC, d_id);

[patch 14/15] zfcp: Reference counting for cfdc requests

[Christof Schmitt]

[-- Attachment #1: adapter-ref.diff --]
[-- Type: text/plain, Size: 996 bytes --]

From: Christof Schmitt <christof.schmitt@de.ibm.com>

Before dropping the reference count with zfcp_adapter_put, increase it
with zfcp_adapter_get when issuing cfdc requests. 

Reviewed-by: Martin Petermann <martin@linux.vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---
 drivers/s390/scsi/zfcp_cfdc.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/s390/scsi/zfcp_cfdc.c	2009-04-17 15:03:18.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_cfdc.c	2009-04-17 15:03:56.000000000 +0200
@@ -4,7 +4,7 @@
  * Userspace interface for accessing the
  * Access Control Lists / Control File Data Channel
  *
- * Copyright IBM Corporation 2008
+ * Copyright IBM Corporation 2008, 2009
  */
 
 #define KMSG_COMPONENT "zfcp"
@@ -197,6 +197,7 @@ static long zfcp_cfdc_dev_ioctl(struct f
 		retval = -ENXIO;
 		goto free_buffer;
 	}
+	zfcp_adapter_get(adapter);
 
 	retval = zfcp_cfdc_sg_setup(data->command, fsf_cfdc->sg,
 				    data_user->control_file);

[patch 15/15] zfcp: Fix oops when port disappears

[Christof Schmitt]

[-- Attachment #1: s390-sles11-12-05-zfcp_fcp_port_handling.patch --]
[-- Type: text/plain, Size: 3238 bytes --]

From: Christof Schmitt <christof.schmitt@de.ibm.com>

The zfcp_port might have been removed, while the FC fast_io_fail timer
is still running and could trigger the terminate_rport_io callback.
Set the pointer to the zfcp_port to NULL and check accordingly
before using it.

Reviewed-by: Martin Petermann <martin@linux.vnet.ibm.com>
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
---
 drivers/s390/scsi/zfcp_aux.c  |    3 +--
 drivers/s390/scsi/zfcp_fsf.c  |    4 ++++
 drivers/s390/scsi/zfcp_scsi.c |   25 +++++++++++++++++++------
 3 files changed, 24 insertions(+), 8 deletions(-)

--- a/drivers/s390/scsi/zfcp_aux.c	2009-04-17 15:03:37.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_aux.c	2009-04-17 15:04:00.000000000 +0200
@@ -671,8 +671,7 @@ void zfcp_port_dequeue(struct zfcp_port 
 	list_del(&port->list);
 	write_unlock_irq(&zfcp_data.config_lock);
 	if (port->rport)
-		fc_remote_port_delete(port->rport);
-	port->rport = NULL;
+		port->rport->dd_data = NULL;
 	zfcp_adapter_put(port->adapter);
 	sysfs_remove_group(&port->sysfs_device.kobj, &zfcp_sysfs_port_attrs);
 	device_unregister(&port->sysfs_device);
--- a/drivers/s390/scsi/zfcp_scsi.c	2009-04-17 15:03:49.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_scsi.c	2009-04-17 15:04:00.000000000 +0200
@@ -486,10 +486,12 @@ static void zfcp_set_rport_dev_loss_tmo(
  */
 static void zfcp_scsi_dev_loss_tmo_callbk(struct fc_rport *rport)
 {
-	struct zfcp_port *port = rport->dd_data;
+	struct zfcp_port *port;
 
 	write_lock_irq(&zfcp_data.config_lock);
-	port->rport = NULL;
+	port = rport->dd_data;
+	if (port)
+		port->rport = NULL;
 	write_unlock_irq(&zfcp_data.config_lock);
 }
 
@@ -503,9 +505,18 @@ static void zfcp_scsi_dev_loss_tmo_callb
  */
 static void zfcp_scsi_terminate_rport_io(struct fc_rport *rport)
 {
-	struct zfcp_port *port = rport->dd_data;
+	struct zfcp_port *port;
+
+	write_lock_irq(&zfcp_data.config_lock);
+	port = rport->dd_data;
+	if (port)
+		zfcp_port_get(port);
+	write_unlock_irq(&zfcp_data.config_lock);
 
-	zfcp_erp_port_reopen(port, 0, "sctrpi1", NULL);
+	if (port) {
+		zfcp_erp_port_reopen(port, 0, "sctrpi1", NULL);
+		zfcp_port_put(port);
+	}
 }
 
 static void zfcp_scsi_rport_register(struct zfcp_port *port)
@@ -534,8 +545,10 @@ static void zfcp_scsi_rport_register(str
 
 static void zfcp_scsi_rport_block(struct zfcp_port *port)
 {
-	if (port->rport)
-		fc_remote_port_delete(port->rport);
+	struct fc_rport *rport = port->rport;
+
+	if (rport)
+		fc_remote_port_delete(rport);
 }
 
 void zfcp_scsi_schedule_rport_register(struct zfcp_port *port)
--- a/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:03:52.000000000 +0200
+++ b/drivers/s390/scsi/zfcp_fsf.c	2009-04-17 15:04:00.000000000 +0200
@@ -172,12 +172,16 @@ static void zfcp_fsf_link_down_info_eval
 					 struct fsf_link_down_info *link_down)
 {
 	struct zfcp_adapter *adapter = req->adapter;
+	unsigned long flags;
 
 	if (atomic_read(&adapter->status) & ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED)
 		return;
 
 	atomic_set_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, &adapter->status);
+
+	read_lock_irqsave(&zfcp_data.config_lock, flags);
 	zfcp_scsi_schedule_rports_block(adapter);
+	read_unlock_irqrestore(&zfcp_data.config_lock, flags);
 
 	if (!link_down)
 		goto out;