Merge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband

Pull infiniband updates from Roland Dreier:
 "Main batch of InfiniBand/RDMA changes for 3.14:
   - Flow steering for InfiniBand UD traffic
   - IP-based addressing for IBoE aka RoCE
   - Pass SRP submaintainership from Dave to Bart
   - SRP transport fixes from Bart
   - Add the new Cisco usNIC low-level device driver
   - Various other fixes"

* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (75 commits)
  IB/mlx5: Verify reserved fields are cleared
  IB/mlx5: Remove old field for create mkey mailbox
  IB/mlx5: Abort driver cleanup if teardown hca fails
  IB/mlx5: Allow creation of QPs with zero-length work queues
  mlx5_core: Fix PowerPC support
  mlx5_core: Improve debugfs readability
  IB/mlx5: Add support for resize CQ
  IB/mlx5: Implement modify CQ
  IB/mlx5: Make sure doorbell record is visible before doorbell
  mlx5_core: Use mlx5 core style warning
  IB/mlx5: Clear out struct before create QP command
  mlx5_core: Fix out arg size in access_register command
  RDMA/nes: Slight optimization of Ethernet address compare
  IB/qib: Fix QP check when looping back to/from QP1
  RDMA/cxgb4: Fix gcc warning on 32-bit arch
  IB/usnic: Remove unused includes of <linux/version.h>
  RDMA/amso1100: Add check if cache memory was allocated before freeing it
  IPoIB: Report operstate consistently when brought up without a link
  IB/core: Fix unused variable warning
  RDMA/cma: Handle global/non-linklocal IPv6 addresses in cma_check_linklocal()
  ...

1 files changed, 57 insertions, 38 deletions

diff --git a/drivers/scsi/scsi_transport_srp.c b/drivers/scsi/scsi_transport_srp.c
index 2700a5a..d47ffc8 100644
--- a/drivers/scsi/scsi_transport_srp.c
+++ b/drivers/scsi/scsi_transport_srp.c
@@ -64,10 +64,14 @@ static inline struct Scsi_Host *rport_to_shost(struct srp_rport *r)
 
 /**
  * srp_tmo_valid() - check timeout combination validity
+ * @reconnect_delay: Reconnect delay in seconds.
+ * @fast_io_fail_tmo: Fast I/O fail timeout in seconds.
+ * @dev_loss_tmo: Device loss timeout in seconds.
  *
  * The combination of the timeout parameters must be such that SCSI commands
  * are finished in a reasonable time. Hence do not allow the fast I/O fail
- * timeout to exceed SCSI_DEVICE_BLOCK_MAX_TIMEOUT. Furthermore, these
+ * timeout to exceed SCSI_DEVICE_BLOCK_MAX_TIMEOUT nor allow dev_loss_tmo to
+ * exceed that limit if failing I/O fast has been disabled. Furthermore, these
  * parameters must be such that multipath can detect failed paths timely.
  * Hence do not allow all three parameters to be disabled simultaneously.
  */
@@ -79,6 +83,9 @@ int srp_tmo_valid(int reconnect_delay, int fast_io_fail_tmo, int dev_loss_tmo)
 		return -EINVAL;
 	if (fast_io_fail_tmo > SCSI_DEVICE_BLOCK_MAX_TIMEOUT)
 		return -EINVAL;
+	if (fast_io_fail_tmo < 0 &&
+	    dev_loss_tmo > SCSI_DEVICE_BLOCK_MAX_TIMEOUT)
+		return -EINVAL;
 	if (dev_loss_tmo >= LONG_MAX / HZ)
 		return -EINVAL;
 	if (fast_io_fail_tmo >= 0 && dev_loss_tmo >= 0 &&
@@ -368,6 +375,7 @@ invalid:
 
 /**
  * srp_reconnect_work() - reconnect and schedule a new attempt if necessary
+ * @work: Work structure used for scheduling this operation.
  */
 static void srp_reconnect_work(struct work_struct *work)
 {
@@ -408,6 +416,7 @@ static void __rport_fail_io_fast(struct srp_rport *rport)
 
 /**
  * rport_fast_io_fail_timedout() - fast I/O failure timeout handler
+ * @work: Work structure used for scheduling this operation.
  */
 static void rport_fast_io_fail_timedout(struct work_struct *work)
 {
@@ -426,6 +435,7 @@ static void rport_fast_io_fail_timedout(struct work_struct *work)
 
 /**
  * rport_dev_loss_timedout() - device loss timeout handler
+ * @work: Work structure used for scheduling this operation.
  */
 static void rport_dev_loss_timedout(struct work_struct *work)
 {
@@ -452,42 +462,35 @@ static void __srp_start_tl_fail_timers(struct srp_rport *rport)
 
 	lockdep_assert_held(&rport->mutex);
 
-	if (!rport->deleted) {
-		delay = rport->reconnect_delay;
-		fast_io_fail_tmo = rport->fast_io_fail_tmo;
-		dev_loss_tmo = rport->dev_loss_tmo;
-		pr_debug("%s current state: %d\n",
-			 dev_name(&shost->shost_gendev), rport->state);
+	delay = rport->reconnect_delay;
+	fast_io_fail_tmo = rport->fast_io_fail_tmo;
+	dev_loss_tmo = rport->dev_loss_tmo;
+	pr_debug("%s current state: %d\n", dev_name(&shost->shost_gendev),
+		 rport->state);
 
-		if (delay > 0)
-			queue_delayed_work(system_long_wq,
-					   &rport->reconnect_work,
-					   1UL * delay * HZ);
-		if (fast_io_fail_tmo >= 0 &&
-		    srp_rport_set_state(rport, SRP_RPORT_BLOCKED) == 0) {
-			pr_debug("%s new state: %d\n",
-				 dev_name(&shost->shost_gendev),
-				 rport->state);
-			scsi_target_block(&shost->shost_gendev);
+	if (rport->state == SRP_RPORT_LOST)
+		return;
+	if (delay > 0)
+		queue_delayed_work(system_long_wq, &rport->reconnect_work,
+				   1UL * delay * HZ);
+	if (srp_rport_set_state(rport, SRP_RPORT_BLOCKED) == 0) {
+		pr_debug("%s new state: %d\n", dev_name(&shost->shost_gendev),
+			 rport->state);
+		scsi_target_block(&shost->shost_gendev);
+		if (fast_io_fail_tmo >= 0)
 			queue_delayed_work(system_long_wq,
 					   &rport->fast_io_fail_work,
 					   1UL * fast_io_fail_tmo * HZ);
-		}
 		if (dev_loss_tmo >= 0)
 			queue_delayed_work(system_long_wq,
 					   &rport->dev_loss_work,
 					   1UL * dev_loss_tmo * HZ);
-	} else {
-		pr_debug("%s has already been deleted\n",
-			 dev_name(&shost->shost_gendev));
-		srp_rport_set_state(rport, SRP_RPORT_FAIL_FAST);
-		scsi_target_unblock(&shost->shost_gendev,
-				    SDEV_TRANSPORT_OFFLINE);
 	}
 }
 
 /**
  * srp_start_tl_fail_timers() - start the transport layer failure timers
+ * @rport: SRP target port.
  *
  * Start the transport layer fast I/O failure and device loss timers. Do not
  * modify a timer that was already started.
@@ -502,6 +505,7 @@ EXPORT_SYMBOL(srp_start_tl_fail_timers);
 
 /**
  * scsi_request_fn_active() - number of kernel threads inside scsi_request_fn()
+ * @shost: SCSI host for which to count the number of scsi_request_fn() callers.
  */
 static int scsi_request_fn_active(struct Scsi_Host *shost)
 {
@@ -522,6 +526,7 @@ static int scsi_request_fn_active(struct Scsi_Host *shost)
 
 /**
  * srp_reconnect_rport() - reconnect to an SRP target port
+ * @rport: SRP target port.
  *
  * Blocks SCSI command queueing before invoking reconnect() such that
  * queuecommand() won't be invoked concurrently with reconnect() from outside
@@ -556,7 +561,7 @@ int srp_reconnect_rport(struct srp_rport *rport)
 	scsi_target_block(&shost->shost_gendev);
 	while (scsi_request_fn_active(shost))
 		msleep(20);
-	res = i->f->reconnect(rport);
+	res = rport->state != SRP_RPORT_LOST ? i->f->reconnect(rport) : -ENODEV;
 	pr_debug("%s (state %d): transport.reconnect() returned %d\n",
 		 dev_name(&shost->shost_gendev), rport->state, res);
 	if (res == 0) {
@@ -578,9 +583,9 @@ int srp_reconnect_rport(struct srp_rport *rport)
 		spin_unlock_irq(shost->host_lock);
 	} else if (rport->state == SRP_RPORT_RUNNING) {
 		/*
-		 * srp_reconnect_rport() was invoked with fast_io_fail
-		 * off. Mark the port as failed and start the TL failure
-		 * timers if these had not yet been started.
+		 * srp_reconnect_rport() has been invoked with fast_io_fail
+		 * and dev_loss off. Mark the port as failed and start the TL
+		 * failure timers if these had not yet been started.
 		 */
 		__rport_fail_io_fast(rport);
 		scsi_target_unblock(&shost->shost_gendev,
@@ -599,6 +604,7 @@ EXPORT_SYMBOL(srp_reconnect_rport);
 
 /**
  * srp_timed_out() - SRP transport intercept of the SCSI timeout EH
+ * @scmd: SCSI command.
  *
  * If a timeout occurs while an rport is in the blocked state, ask the SCSI
  * EH to continue waiting (BLK_EH_RESET_TIMER). Otherwise let the SCSI core
@@ -622,10 +628,6 @@ static void srp_rport_release(struct device *dev)
 {
 	struct srp_rport *rport = dev_to_rport(dev);
 
-	cancel_delayed_work_sync(&rport->reconnect_work);
-	cancel_delayed_work_sync(&rport->fast_io_fail_work);
-	cancel_delayed_work_sync(&rport->dev_loss_work);
-
 	put_device(dev->parent);
 	kfree(rport);
 }
@@ -674,6 +676,7 @@ static int srp_host_match(struct attribute_container *cont, struct device *dev)
 
 /**
  * srp_rport_get() - increment rport reference count
+ * @rport: SRP target port.
  */
 void srp_rport_get(struct srp_rport *rport)
 {
@@ -683,6 +686,7 @@ EXPORT_SYMBOL(srp_rport_get);
 
 /**
  * srp_rport_put() - decrement rport reference count
+ * @rport: SRP target port.
  */
 void srp_rport_put(struct srp_rport *rport)
 {
@@ -780,12 +784,6 @@ void srp_rport_del(struct srp_rport *rport)
 	device_del(dev);
 	transport_destroy_device(dev);
 
-	mutex_lock(&rport->mutex);
-	if (rport->state == SRP_RPORT_BLOCKED)
-		__rport_fail_io_fast(rport);
-	rport->deleted = true;
-	mutex_unlock(&rport->mutex);
-
 	put_device(dev);
 }
 EXPORT_SYMBOL_GPL(srp_rport_del);
@@ -810,6 +808,27 @@ void srp_remove_host(struct Scsi_Host *shost)
 }
 EXPORT_SYMBOL_GPL(srp_remove_host);
 
+/**
+ * srp_stop_rport_timers - stop the transport layer recovery timers
+ *
+ * Must be called after srp_remove_host() and scsi_remove_host(). The caller
+ * must hold a reference on the rport (rport->dev) and on the SCSI host
+ * (rport->dev.parent).
+ */
+void srp_stop_rport_timers(struct srp_rport *rport)
+{
+	mutex_lock(&rport->mutex);
+	if (rport->state == SRP_RPORT_BLOCKED)
+		__rport_fail_io_fast(rport);
+	srp_rport_set_state(rport, SRP_RPORT_LOST);
+	mutex_unlock(&rport->mutex);
+
+	cancel_delayed_work_sync(&rport->reconnect_work);
+	cancel_delayed_work_sync(&rport->fast_io_fail_work);
+	cancel_delayed_work_sync(&rport->dev_loss_work);
+}
+EXPORT_SYMBOL_GPL(srp_stop_rport_timers);
+
 static int srp_tsk_mgmt_response(struct Scsi_Host *shost, u64 nexus, u64 tm_id,
 				 int result)
 {