[patch 0/6] zfcp: updates, fixing a variety of oddities

[patch 0/6] zfcp: updates, fixing a variety of oddities

[swen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1945 bytes --]

-- 
patch 1/6: Remove unnecessary eh_bus_reset_handler callsback
           The callback function used by zfcp always returns success,
           which is an indication for the SCSI midlayer to stop error
           handling. Remove the bus_reset callback, since the same
           function will be called via the host_reset callback.

patch 2/6: Use also port and adapter to identify unit in messages
           Add the appropriate messages.

patch 3/6: Specify waiting times in ERP in seconds
           It is not necessary to use jiffies or milliseconds to specify
           waiting times that last a couple of seconds.

patch 4/6: Remove SCSI devices when removing adapter
           The common I/O layer can call remove a handler to inform zfcp
           that a device disappeared. The handler zfcp_ccw_remove then
           removes all unit, port and the adapter data structures. Removing
           the units requires that the SCSI devices are removed first.


patch 5/6: Fix deadlock when adding invalid LUN
           When adding an invalid LUN, there is a deadlock between the add
           via scsi_scan_target and the slave_destroy handler: The handler
           waits for the scan to complete, but for an invalid unit,
           scsi_scan_target directly calls the slave_destroy handler.
           Fix the deadlock by removing the wait in the slave_destroy
           handler, it was not necessary anyway.

patch 6/6: Reduce flood on HBA trace
           Remove tracing for request with a "qualifier" field set in the
           response.  The protocol status qualifier now contains measurement
           data for "good" commands, so this check would trace every response
           by default.
           The fix is to simply remove the "qual" tracing: The responses with an
           interesting status are also traced as "ferr" or "perr" and all
           responses can be traced as "norm" with a higher trace level.

[patch 1/6] zfcp: Remove unnecessary eh_bus_reset_handler callback

[swen]

[-- Attachment #1: 800-zfcp-callback.diff --]
[-- Type: text/plain, Size: 1703 bytes --]

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

The callback function used by zfcp always returns success,
which is an indication for the SCSI midlayer to stop error
handling. Remove the bus_reset callback, since the same
function will be called via the host_reset callback.

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

 drivers/s390/scsi/zfcp_scsi.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

Index: scsi-misc/drivers/s390/scsi/zfcp_scsi.c
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_scsi.c
+++ scsi-misc/drivers/s390/scsi/zfcp_scsi.c
@@ -51,7 +51,6 @@ struct zfcp_data zfcp_data = {
 		.queuecommand		= zfcp_scsi_queuecommand,
 		.eh_abort_handler	= zfcp_scsi_eh_abort_handler,
 		.eh_device_reset_handler = zfcp_scsi_eh_device_reset_handler,
-		.eh_bus_reset_handler	= zfcp_scsi_eh_host_reset_handler,
 		.eh_host_reset_handler	= zfcp_scsi_eh_host_reset_handler,
 		.can_queue		= 4096,
 		.this_id		= -1,
@@ -542,7 +541,7 @@ zfcp_task_management_function(struct zfc
 }
 
 /**
- * zfcp_scsi_eh_host_reset_handler - handler for host and bus reset
+ * zfcp_scsi_eh_host_reset_handler - handler for host reset
  */
 static int zfcp_scsi_eh_host_reset_handler(struct scsi_cmnd *scpnt)
 {
@@ -552,7 +551,7 @@ static int zfcp_scsi_eh_host_reset_handl
 	unit = (struct zfcp_unit*) scpnt->device->hostdata;
 	adapter = unit->port->adapter;
 
-	ZFCP_LOG_NORMAL("host/bus reset because of problems with "
+	ZFCP_LOG_NORMAL("host reset because of problems with "
 			"unit 0x%016Lx\n", unit->fcp_lun);
 
 	zfcp_erp_adapter_reopen(adapter, 0);

-- 

[patch 2/6] zfcp: Use also port and adapter to identify unit in messages.

[swen]

[-- Attachment #1: 802-zfcp-msg.diff --]
[-- Type: text/plain, Size: 1342 bytes --]

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

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

 drivers/s390/scsi/zfcp_scsi.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff -urpN linux-2.6/drivers/s390/scsi/zfcp_scsi.c linux-2.6-patched/drivers/s390/scsi/zfcp_scsi.c
--- linux-2.6/drivers/s390/scsi/zfcp_scsi.c	2007-11-02 08:38:25.000000000 +0100
+++ linux-2.6-patched/drivers/s390/scsi/zfcp_scsi.c	2007-11-02 08:38:25.000000000 +0100
@@ -458,7 +458,9 @@ zfcp_scsi_eh_device_reset_handler(struct
 		retval = SUCCESS;
 		goto out;
 	}
-	ZFCP_LOG_NORMAL("resetting unit 0x%016Lx\n", unit->fcp_lun);
+	ZFCP_LOG_NORMAL("resetting unit 0x%016Lx on port 0x%016Lx, adapter %s\n",
+			unit->fcp_lun, unit->port->wwpn,
+			zfcp_get_busid_by_adapter(unit->port->adapter));
 
 	/*
 	 * If we do not know whether the unit supports 'logical unit reset'
@@ -552,7 +554,9 @@ static int zfcp_scsi_eh_host_reset_handl
 	adapter = unit->port->adapter;
 
 	ZFCP_LOG_NORMAL("host reset because of problems with "
-			"unit 0x%016Lx\n", unit->fcp_lun);
+		"unit 0x%016Lx on port 0x%016Lx, adapter %s\n",
+		unit->fcp_lun, unit->port->wwpn,
+		zfcp_get_busid_by_adapter(unit->port->adapter));
 
 	zfcp_erp_adapter_reopen(adapter, 0);
 	zfcp_erp_wait(adapter);

-- 

[patch 3/6] zfcp: Specify waiting times in ERP in seconds

[swen]

[-- Attachment #1: 803-zfcp-waits.diff --]
[-- Type: text/plain, Size: 2896 bytes --]

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

It is not necessary to use jiffies or milliseconds to specify
waiting times that last a couple of seconds.

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

 drivers/s390/scsi/zfcp_def.h |    4 ++--
 drivers/s390/scsi/zfcp_erp.c |    8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

Index: scsi-misc/drivers/s390/scsi/zfcp_def.h
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_def.h
+++ scsi-misc/drivers/s390/scsi/zfcp_def.h
@@ -117,7 +117,7 @@ zfcp_address_to_sg(void *address, struct
 
 #define ZFCP_SBAL_TIMEOUT               (5*HZ)
 
-#define ZFCP_TYPE2_RECOVERY_TIME        (8*HZ)
+#define ZFCP_TYPE2_RECOVERY_TIME        8	/* seconds */
 
 /* queue polling (values in microseconds) */
 #define ZFCP_MAX_INPUT_THRESHOLD 	5000	/* FIXME: tune */
@@ -138,7 +138,7 @@ zfcp_address_to_sg(void *address, struct
 #define ZFCP_STATUS_READS_RECOM		        FSF_STATUS_READS_RECOM
 
 /* Do 1st retry in 1 second, then double the timeout for each following retry */
-#define ZFCP_EXCHANGE_CONFIG_DATA_FIRST_SLEEP	100
+#define ZFCP_EXCHANGE_CONFIG_DATA_FIRST_SLEEP	1
 #define ZFCP_EXCHANGE_CONFIG_DATA_RETRIES	7
 
 /* timeout value for "default timer" for fsf requests */
Index: scsi-misc/drivers/s390/scsi/zfcp_erp.c
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_erp.c
+++ scsi-misc/drivers/s390/scsi/zfcp_erp.c
@@ -131,7 +131,7 @@ static void zfcp_close_qdio(struct zfcp_
 	debug_text_event(adapter->erp_dbf, 3, "qdio_down2a");
 	while (qdio_shutdown(adapter->ccw_device,
 			     QDIO_FLAG_CLEANUP_USING_CLEAR) == -EINPROGRESS)
-		msleep(1000);
+		ssleep(1);
 	debug_text_event(adapter->erp_dbf, 3, "qdio_down2b");
 
 	/* cleanup used outbound sbals */
@@ -1900,7 +1900,7 @@ zfcp_erp_adapter_strategy(struct zfcp_er
 		ZFCP_LOG_INFO("Waiting to allow the adapter %s "
 			      "to recover itself\n",
 			      zfcp_get_busid_by_adapter(adapter));
-		msleep(jiffies_to_msecs(ZFCP_TYPE2_RECOVERY_TIME));
+		ssleep(ZFCP_TYPE2_RECOVERY_TIME);
 	}
 
 	return retval;
@@ -2080,7 +2080,7 @@ zfcp_erp_adapter_strategy_open_qdio(stru
 	debug_text_event(adapter->erp_dbf, 3, "qdio_down1a");
 	while (qdio_shutdown(adapter->ccw_device,
 			     QDIO_FLAG_CLEANUP_USING_CLEAR) == -EINPROGRESS)
-		msleep(1000);
+		ssleep(1);
 	debug_text_event(adapter->erp_dbf, 3, "qdio_down1b");
 
  failed_qdio_establish:
@@ -2165,7 +2165,7 @@ zfcp_erp_adapter_strategy_open_fsf_xconf
 		ZFCP_LOG_DEBUG("host connection still initialising... "
 			       "waiting and retrying...\n");
 		/* sleep a little bit before retry */
-		msleep(jiffies_to_msecs(sleep));
+		ssleep(sleep);
 		sleep *= 2;
 	}
 

-- 

[patch 4/6] zfcp: Remove SCSI devices when removing complete adapter

[swen]

[-- Attachment #1: 804-zfcp-rmdevs.diff --]
[-- Type: text/plain, Size: 1138 bytes --]

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

The common I/O layer can call remove a handler to inform zfcp
that a device disappeared. The handler zfcp_ccw_remove then
removes all unit, port and the adapter data structures. Removing
the units requires that the SCSI devices are removed first.

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

 drivers/s390/scsi/zfcp_ccw.c |    3 +++
 1 file changed, 3 insertions(+)

diff -urpN linux-2.6/drivers/s390/scsi/zfcp_ccw.c linux-2.6-patched/drivers/s390/scsi/zfcp_ccw.c
--- linux-2.6/drivers/s390/scsi/zfcp_ccw.c	2007-11-02 08:37:56.000000000 +0100
+++ linux-2.6-patched/drivers/s390/scsi/zfcp_ccw.c	2007-11-02 08:38:26.000000000 +0100
@@ -120,6 +120,9 @@ zfcp_ccw_remove(struct ccw_device *ccw_d
 
 	list_for_each_entry_safe(port, p, &adapter->port_remove_lh, list) {
 		list_for_each_entry_safe(unit, u, &port->unit_remove_lh, list) {
+			if (atomic_test_mask(ZFCP_STATUS_UNIT_REGISTERED,
+				&unit->status))
+				scsi_remove_device(unit->device);
 			zfcp_unit_dequeue(unit);
 		}
 		zfcp_port_dequeue(port);

-- 

[patch 5/6] zfcp: Fix deadlock when adding invalid LUN

[swen]

[-- Attachment #1: 805-zfcp-deadlock.diff --]
[-- Type: text/plain, Size: 3007 bytes --]

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

When adding an invalid LUN, there is a deadlock between the add
via scsi_scan_target and the slave_destroy handler: The handler
waits for the scan to complete, but for an invalid unit,
scsi_scan_target directly calls the slave_destroy handler.

Fix the deadlock by removing the wait in the slave_destroy
handler, it was not necessary anyway.

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

 drivers/s390/scsi/zfcp_aux.c  |    2 --
 drivers/s390/scsi/zfcp_def.h  |    4 ----
 drivers/s390/scsi/zfcp_erp.c  |    1 -
 drivers/s390/scsi/zfcp_scsi.c |    3 ---
 4 files changed, 10 deletions(-)

Index: scsi-misc/drivers/s390/scsi/zfcp_aux.c
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_aux.c
+++ scsi-misc/drivers/s390/scsi/zfcp_aux.c
@@ -844,8 +844,6 @@ zfcp_unit_enqueue(struct zfcp_port *port
 	unit->sysfs_device.release = zfcp_sysfs_unit_release;
 	dev_set_drvdata(&unit->sysfs_device, unit);
 
-	init_waitqueue_head(&unit->scsi_scan_wq);
-
 	/* mark unit unusable as long as sysfs registration is not complete */
 	atomic_set_mask(ZFCP_STATUS_COMMON_REMOVE, &unit->status);
 
Index: scsi-misc/drivers/s390/scsi/zfcp_def.h
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_def.h
+++ scsi-misc/drivers/s390/scsi/zfcp_def.h
@@ -982,10 +982,6 @@ struct zfcp_unit {
         struct scsi_device     *device;        /* scsi device struct pointer */
 	struct zfcp_erp_action erp_action;     /* pending error recovery */
         atomic_t               erp_counter;
-	wait_queue_head_t      scsi_scan_wq;   /* can be used to wait until
-						  all scsi_scan_target
-						  requests have been
-						  completed. */
 };
 
 /* FSF request */
Index: scsi-misc/drivers/s390/scsi/zfcp_erp.c
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_erp.c
+++ scsi-misc/drivers/s390/scsi/zfcp_erp.c
@@ -1609,7 +1609,6 @@ static void zfcp_erp_scsi_scan(struct wo
 	scsi_scan_target(&rport->dev, 0, rport->scsi_target_id,
 			 unit->scsi_lun, 0);
 	atomic_clear_mask(ZFCP_STATUS_UNIT_SCSI_WORK_PENDING, &unit->status);
-	wake_up(&unit->scsi_scan_wq);
 	zfcp_unit_put(unit);
 	kfree(p);
 }
Index: scsi-misc/drivers/s390/scsi/zfcp_scsi.c
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_scsi.c
+++ scsi-misc/drivers/s390/scsi/zfcp_scsi.c
@@ -180,9 +180,6 @@ static void zfcp_scsi_slave_destroy(stru
 
 	if (unit) {
 		zfcp_erp_wait(unit->port->adapter);
-		wait_event(unit->scsi_scan_wq,
-			   atomic_test_mask(ZFCP_STATUS_UNIT_SCSI_WORK_PENDING,
-					    &unit->status) == 0);
 		atomic_clear_mask(ZFCP_STATUS_UNIT_REGISTERED, &unit->status);
 		sdpnt->hostdata = NULL;
 		unit->device = NULL;

-- 

[patch 6/6] zfcp: Reduce flood on hba trace

[swen]

[-- Attachment #1: 806-zfcp-trace.diff --]
[-- Type: text/plain, Size: 1376 bytes --]

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

Remove tracing for request with a "qualifier" field set in the
response.  The protocol status qualifier now contains measurement
data for "good" commands, so this check would trace every response
by default.

The fix is to simply remove the "qual" tracing: The responses with an
interesting status are also traced as "ferr" or "perr" and all
responses can be traced as "norm" with a higher trace level.

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

 drivers/s390/scsi/zfcp_dbf.c |    6 ------
 1 file changed, 6 deletions(-)

Index: scsi-misc/drivers/s390/scsi/zfcp_dbf.c
===================================================================
--- scsi-misc.orig/drivers/s390/scsi/zfcp_dbf.c
+++ scsi-misc/drivers/s390/scsi/zfcp_dbf.c
@@ -161,12 +161,6 @@ void zfcp_hba_dbf_event_fsf_response(str
 		   (fsf_req->fsf_command == FSF_QTCB_OPEN_LUN)) {
 		strncpy(rec->tag2, "open", ZFCP_DBF_TAG_SIZE);
 		level = 4;
-	} else if ((prot_status_qual->doubleword[0] != 0) ||
-		   (prot_status_qual->doubleword[1] != 0) ||
-		   (fsf_status_qual->doubleword[0] != 0) ||
-		   (fsf_status_qual->doubleword[1] != 0)) {
-		strncpy(rec->tag2, "qual", ZFCP_DBF_TAG_SIZE);
-		level = 3;
 	} else {
 		strncpy(rec->tag2, "norm", ZFCP_DBF_TAG_SIZE);
 		level = 6;

--