Merge tag 'gfs2-4.16.rc4.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2

Pull gfs2 fix from Bob Peterson:
 "An additional patch from Andreas Gruenbacher that fixes another
  unfortunate GFS2 regression"

* tag 'gfs2-4.16.rc4.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
  gfs2: Fixes to "Implement iomap for block_map" (2)
diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index 65154ea..6c8ce15 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -63,6 +63,7 @@
 				   _ASCE_USER_BITS | _ASCE_TYPE_SEGMENT;
 		/* pgd_alloc() did not account this pmd */
 		mm_inc_nr_pmds(mm);
+		mm_inc_nr_puds(mm);
 	}
 	crst_table_init((unsigned long *) mm->pgd, pgd_entry_type(mm));
 	return 0;
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 13a133a..a5621ea 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -14,6 +14,7 @@
 #include <asm/processor.h>
 #include <asm/cache.h>
 #include <asm/ctl_reg.h>
+#include <asm/dwarf.h>
 #include <asm/errno.h>
 #include <asm/ptrace.h>
 #include <asm/thread_info.h>
@@ -230,7 +231,7 @@
 	.hidden \name
 	.type \name,@function
 \name:
-	.cfi_startproc
+	CFI_STARTPROC
 #ifdef CONFIG_HAVE_MARCH_Z10_FEATURES
 	exrl	0,0f
 #else
@@ -239,7 +240,7 @@
 #endif
 	j	.
 0:	br	\reg
-	.cfi_endproc
+	CFI_ENDPROC
 	.endm
 
 	GEN_BR_THUNK __s390x_indirect_jump_r1use_r9,%r9,%r1
@@ -426,13 +427,13 @@
 	UPDATE_VTIME %r8,%r9,__LC_SYNC_ENTER_TIMER
 	BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP
 	stmg	%r0,%r7,__PT_R0(%r11)
-	# clear user controlled register to prevent speculative use
-	xgr	%r0,%r0
 	mvc	__PT_R8(64,%r11),__LC_SAVE_AREA_SYNC
 	mvc	__PT_PSW(16,%r11),__LC_SVC_OLD_PSW
 	mvc	__PT_INT_CODE(4,%r11),__LC_SVC_ILC
 	stg	%r14,__PT_FLAGS(%r11)
 .Lsysc_do_svc:
+	# clear user controlled register to prevent speculative use
+	xgr	%r0,%r0
 	# load address of system call table
 	lg	%r10,__THREAD_sysc_table(%r13,%r12)
 	llgh	%r8,__PT_INT_CODE+2(%r11)
@@ -1439,6 +1440,7 @@
 	stg	%r15,__LC_SYSTEM_TIMER
 0:	# update accounting time stamp
 	mvc	__LC_LAST_UPDATE_TIMER(8),__LC_SYNC_ENTER_TIMER
+	BPENTER __TI_flags(%r12),_TIF_ISOLATE_BP
 	# set up saved register r11
 	lg	%r15,__LC_KERNEL_STACK
 	la	%r9,STACK_FRAME_OVERHEAD(%r15)
diff --git a/arch/s390/kernel/nospec-branch.c b/arch/s390/kernel/nospec-branch.c
index 69d7fcf..9aff72d 100644
--- a/arch/s390/kernel/nospec-branch.c
+++ b/arch/s390/kernel/nospec-branch.c
@@ -2,8 +2,8 @@
 #include <linux/module.h>
 #include <asm/nospec-branch.h>
 
-int nospec_call_disable = IS_ENABLED(EXPOLINE_OFF);
-int nospec_return_disable = !IS_ENABLED(EXPOLINE_FULL);
+int nospec_call_disable = IS_ENABLED(CONFIG_EXPOLINE_OFF);
+int nospec_return_disable = !IS_ENABLED(CONFIG_EXPOLINE_FULL);
 
 static int __init nospectre_v2_setup_early(char *str)
 {
diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index dd4708c..1fc0c08 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -4310,7 +4310,7 @@
 
 	rstate = regulator_get_suspend_state(rdev, *state);
 	if (rstate == NULL)
-		return -EINVAL;
+		return 0;
 
 	mutex_lock(&rdev->mutex);
 
diff --git a/drivers/regulator/stm32-vrefbuf.c b/drivers/regulator/stm32-vrefbuf.c
index 72c8b3e..e0a9c44 100644
--- a/drivers/regulator/stm32-vrefbuf.c
+++ b/drivers/regulator/stm32-vrefbuf.c
@@ -51,7 +51,7 @@
 	 * arbitrary timeout.
 	 */
 	ret = readl_poll_timeout(priv->base + STM32_VREFBUF_CSR, val,
-				 !(val & STM32_VRR), 650, 10000);
+				 val & STM32_VRR, 650, 10000);
 	if (ret) {
 		dev_err(&rdev->dev, "stm32 vrefbuf timed out!\n");
 		val = readl_relaxed(priv->base + STM32_VREFBUF_CSR);
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index a7c15f0..ecef8e7 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -2581,8 +2581,6 @@
 	case DASD_CQR_QUEUED:
 		/* request was not started - just set to cleared */
 		cqr->status = DASD_CQR_CLEARED;
-		if (cqr->callback_data == DASD_SLEEPON_START_TAG)
-			cqr->callback_data = DASD_SLEEPON_END_TAG;
 		break;
 	case DASD_CQR_IN_IO:
 		/* request in IO - terminate IO and release again */
@@ -3902,9 +3900,12 @@
 		wait_event(dasd_flush_wq,
 			   (cqr->status != DASD_CQR_CLEAR_PENDING));
 
-		/* mark sleepon requests as ended */
-		if (cqr->callback_data == DASD_SLEEPON_START_TAG)
-			cqr->callback_data = DASD_SLEEPON_END_TAG;
+		/*
+		 * requeue requests to blocklayer will only work
+		 * for block device requests
+		 */
+		if (_dasd_requeue_request(cqr))
+			continue;
 
 		/* remove requests from device and block queue */
 		list_del_init(&cqr->devlist);
@@ -3917,13 +3918,6 @@
 			cqr = refers;
 		}
 
-		/*
-		 * requeue requests to blocklayer will only work
-		 * for block device requests
-		 */
-		if (_dasd_requeue_request(cqr))
-			continue;
-
 		if (cqr->block)
 			list_del_init(&cqr->blocklist);
 		cqr->block->base->discipline->free_cp(
@@ -3940,8 +3934,7 @@
 		list_splice_tail(&requeue_queue, &device->ccw_queue);
 		spin_unlock_irq(get_ccwdev_lock(device->cdev));
 	}
-	/* wake up generic waitqueue for eventually ended sleepon requests */
-	wake_up(&generic_waitq);
+	dasd_schedule_device_bh(device);
 	return rc;
 }
 
diff --git a/drivers/s390/cio/device_fsm.c b/drivers/s390/cio/device_fsm.c
index 1319122..9169af7 100644
--- a/drivers/s390/cio/device_fsm.c
+++ b/drivers/s390/cio/device_fsm.c
@@ -795,6 +795,7 @@
 
 	ccw_device_set_timeout(cdev, 0);
 	cdev->private->iretry = 255;
+	cdev->private->async_kill_io_rc = -ETIMEDOUT;
 	ret = ccw_device_cancel_halt_clear(cdev);
 	if (ret == -EBUSY) {
 		ccw_device_set_timeout(cdev, 3*HZ);
@@ -871,7 +872,7 @@
 	/* OK, i/o is dead now. Call interrupt handler. */
 	if (cdev->handler)
 		cdev->handler(cdev, cdev->private->intparm,
-			      ERR_PTR(-EIO));
+			      ERR_PTR(cdev->private->async_kill_io_rc));
 }
 
 static void
@@ -888,14 +889,16 @@
 	ccw_device_online_verify(cdev, 0);
 	if (cdev->handler)
 		cdev->handler(cdev, cdev->private->intparm,
-			      ERR_PTR(-EIO));
+			      ERR_PTR(cdev->private->async_kill_io_rc));
 }
 
 void ccw_device_kill_io(struct ccw_device *cdev)
 {
 	int ret;
 
+	ccw_device_set_timeout(cdev, 0);
 	cdev->private->iretry = 255;
+	cdev->private->async_kill_io_rc = -EIO;
 	ret = ccw_device_cancel_halt_clear(cdev);
 	if (ret == -EBUSY) {
 		ccw_device_set_timeout(cdev, 3*HZ);
diff --git a/drivers/s390/cio/device_ops.c b/drivers/s390/cio/device_ops.c
index 1caf6a3..75ce12a 100644
--- a/drivers/s390/cio/device_ops.c
+++ b/drivers/s390/cio/device_ops.c
@@ -159,7 +159,7 @@
 }
 
 /**
- * ccw_device_start_key() - start a s390 channel program with key
+ * ccw_device_start_timeout_key() - start a s390 channel program with timeout and key
  * @cdev: target ccw device
  * @cpa: logical start address of channel program
  * @intparm: user specific interruption parameter; will be presented back to
@@ -170,10 +170,15 @@
  * @key: storage key to be used for the I/O
  * @flags: additional flags; defines the action to be performed for I/O
  *	   processing.
+ * @expires: timeout value in jiffies
  *
  * Start a S/390 channel program. When the interrupt arrives, the
  * IRQ handler is called, either immediately, delayed (dev-end missing,
  * or sense required) or never (no IRQ handler registered).
+ * This function notifies the device driver if the channel program has not
+ * completed during the time specified by @expires. If a timeout occurs, the
+ * channel program is terminated via xsch, hsch or csch, and the device's
+ * interrupt handler will be called with an irb containing ERR_PTR(-%ETIMEDOUT).
  * Returns:
  *  %0, if the operation was successful;
  *  -%EBUSY, if the device is busy, or status pending;
@@ -182,9 +187,9 @@
  * Context:
  *  Interrupts disabled, ccw device lock held
  */
-int ccw_device_start_key(struct ccw_device *cdev, struct ccw1 *cpa,
-			 unsigned long intparm, __u8 lpm, __u8 key,
-			 unsigned long flags)
+int ccw_device_start_timeout_key(struct ccw_device *cdev, struct ccw1 *cpa,
+				 unsigned long intparm, __u8 lpm, __u8 key,
+				 unsigned long flags, int expires)
 {
 	struct subchannel *sch;
 	int ret;
@@ -224,6 +229,8 @@
 	switch (ret) {
 	case 0:
 		cdev->private->intparm = intparm;
+		if (expires)
+			ccw_device_set_timeout(cdev, expires);
 		break;
 	case -EACCES:
 	case -ENODEV:
@@ -234,7 +241,7 @@
 }
 
 /**
- * ccw_device_start_timeout_key() - start a s390 channel program with timeout and key
+ * ccw_device_start_key() - start a s390 channel program with key
  * @cdev: target ccw device
  * @cpa: logical start address of channel program
  * @intparm: user specific interruption parameter; will be presented back to
@@ -245,15 +252,10 @@
  * @key: storage key to be used for the I/O
  * @flags: additional flags; defines the action to be performed for I/O
  *	   processing.
- * @expires: timeout value in jiffies
  *
  * Start a S/390 channel program. When the interrupt arrives, the
  * IRQ handler is called, either immediately, delayed (dev-end missing,
  * or sense required) or never (no IRQ handler registered).
- * This function notifies the device driver if the channel program has not
- * completed during the time specified by @expires. If a timeout occurs, the
- * channel program is terminated via xsch, hsch or csch, and the device's
- * interrupt handler will be called with an irb containing ERR_PTR(-%ETIMEDOUT).
  * Returns:
  *  %0, if the operation was successful;
  *  -%EBUSY, if the device is busy, or status pending;
@@ -262,19 +264,12 @@
  * Context:
  *  Interrupts disabled, ccw device lock held
  */
-int ccw_device_start_timeout_key(struct ccw_device *cdev, struct ccw1 *cpa,
-				 unsigned long intparm, __u8 lpm, __u8 key,
-				 unsigned long flags, int expires)
+int ccw_device_start_key(struct ccw_device *cdev, struct ccw1 *cpa,
+			 unsigned long intparm, __u8 lpm, __u8 key,
+			 unsigned long flags)
 {
-	int ret;
-
-	if (!cdev)
-		return -ENODEV;
-	ccw_device_set_timeout(cdev, expires);
-	ret = ccw_device_start_key(cdev, cpa, intparm, lpm, key, flags);
-	if (ret != 0)
-		ccw_device_set_timeout(cdev, 0);
-	return ret;
+	return ccw_device_start_timeout_key(cdev, cpa, intparm, lpm, key,
+					    flags, 0);
 }
 
 /**
@@ -489,18 +484,20 @@
 EXPORT_SYMBOL(ccw_device_get_id);
 
 /**
- * ccw_device_tm_start_key() - perform start function
+ * ccw_device_tm_start_timeout_key() - perform start function
  * @cdev: ccw device on which to perform the start function
  * @tcw: transport-command word to be started
  * @intparm: user defined parameter to be passed to the interrupt handler
  * @lpm: mask of paths to use
  * @key: storage key to use for storage access
+ * @expires: time span in jiffies after which to abort request
  *
  * Start the tcw on the given ccw device. Return zero on success, non-zero
  * otherwise.
  */
-int ccw_device_tm_start_key(struct ccw_device *cdev, struct tcw *tcw,
-			    unsigned long intparm, u8 lpm, u8 key)
+int ccw_device_tm_start_timeout_key(struct ccw_device *cdev, struct tcw *tcw,
+				    unsigned long intparm, u8 lpm, u8 key,
+				    int expires)
 {
 	struct subchannel *sch;
 	int rc;
@@ -527,37 +524,32 @@
 			return -EACCES;
 	}
 	rc = cio_tm_start_key(sch, tcw, lpm, key);
-	if (rc == 0)
+	if (rc == 0) {
 		cdev->private->intparm = intparm;
+		if (expires)
+			ccw_device_set_timeout(cdev, expires);
+	}
 	return rc;
 }
-EXPORT_SYMBOL(ccw_device_tm_start_key);
+EXPORT_SYMBOL(ccw_device_tm_start_timeout_key);
 
 /**
- * ccw_device_tm_start_timeout_key() - perform start function
+ * ccw_device_tm_start_key() - perform start function
  * @cdev: ccw device on which to perform the start function
  * @tcw: transport-command word to be started
  * @intparm: user defined parameter to be passed to the interrupt handler
  * @lpm: mask of paths to use
  * @key: storage key to use for storage access
- * @expires: time span in jiffies after which to abort request
  *
  * Start the tcw on the given ccw device. Return zero on success, non-zero
  * otherwise.
  */
-int ccw_device_tm_start_timeout_key(struct ccw_device *cdev, struct tcw *tcw,
-				    unsigned long intparm, u8 lpm, u8 key,
-				    int expires)
+int ccw_device_tm_start_key(struct ccw_device *cdev, struct tcw *tcw,
+			    unsigned long intparm, u8 lpm, u8 key)
 {
-	int ret;
-
-	ccw_device_set_timeout(cdev, expires);
-	ret = ccw_device_tm_start_key(cdev, tcw, intparm, lpm, key);
-	if (ret != 0)
-		ccw_device_set_timeout(cdev, 0);
-	return ret;
+	return ccw_device_tm_start_timeout_key(cdev, tcw, intparm, lpm, key, 0);
 }
-EXPORT_SYMBOL(ccw_device_tm_start_timeout_key);
+EXPORT_SYMBOL(ccw_device_tm_start_key);
 
 /**
  * ccw_device_tm_start() - perform start function
diff --git a/drivers/s390/cio/io_sch.h b/drivers/s390/cio/io_sch.h
index af571d8..90e4e3a 100644
--- a/drivers/s390/cio/io_sch.h
+++ b/drivers/s390/cio/io_sch.h
@@ -157,6 +157,7 @@
 	unsigned long intparm;	/* user interruption parameter */
 	struct qdio_irq *qdio_data;
 	struct irb irb;		/* device status */
+	int async_kill_io_rc;
 	struct senseid senseid;	/* SenseID info */
 	struct pgid pgid[8];	/* path group IDs per chpid*/
 	struct ccw1 iccws[2];	/* ccws for SNID/SID/SPGID commands */
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 57bf43e..dd94649 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -328,8 +328,6 @@
 	if (shost->work_q)
 		destroy_workqueue(shost->work_q);
 
-	destroy_rcu_head(&shost->rcu);
-
 	if (shost->shost_state == SHOST_CREATED) {
 		/*
 		 * Free the shost_dev device name here if scsi_host_alloc()
@@ -404,7 +402,6 @@
 	INIT_LIST_HEAD(&shost->starved_list);
 	init_waitqueue_head(&shost->host_wait);
 	mutex_init(&shost->scan_mutex);
-	init_rcu_head(&shost->rcu);
 
 	index = ida_simple_get(&host_index_ida, 0, 0, GFP_KERNEL);
 	if (index < 0)
diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c
index 073ced0..dc8e850 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
@@ -216,36 +216,30 @@
 /**
  * megasas_fire_cmd_fusion -	Sends command to the FW
  * @instance:			Adapter soft state
- * @req_desc:			32bit or 64bit Request descriptor
+ * @req_desc:			64bit Request descriptor
  *
- * Perform PCI Write. Ventura supports 32 bit Descriptor.
- * Prior to Ventura (12G) MR controller supports 64 bit Descriptor.
+ * Perform PCI Write.
  */
 
 static void
 megasas_fire_cmd_fusion(struct megasas_instance *instance,
 		union MEGASAS_REQUEST_DESCRIPTOR_UNION *req_desc)
 {
-	if (instance->adapter_type == VENTURA_SERIES)
-		writel(le32_to_cpu(req_desc->u.low),
-			&instance->reg_set->inbound_single_queue_port);
-	else {
 #if defined(writeq) && defined(CONFIG_64BIT)
-		u64 req_data = (((u64)le32_to_cpu(req_desc->u.high) << 32) |
-				le32_to_cpu(req_desc->u.low));
+	u64 req_data = (((u64)le32_to_cpu(req_desc->u.high) << 32) |
+		le32_to_cpu(req_desc->u.low));
 
-		writeq(req_data, &instance->reg_set->inbound_low_queue_port);
+	writeq(req_data, &instance->reg_set->inbound_low_queue_port);
 #else
-		unsigned long flags;
-		spin_lock_irqsave(&instance->hba_lock, flags);
-		writel(le32_to_cpu(req_desc->u.low),
-			&instance->reg_set->inbound_low_queue_port);
-		writel(le32_to_cpu(req_desc->u.high),
-			&instance->reg_set->inbound_high_queue_port);
-		mmiowb();
-		spin_unlock_irqrestore(&instance->hba_lock, flags);
+	unsigned long flags;
+	spin_lock_irqsave(&instance->hba_lock, flags);
+	writel(le32_to_cpu(req_desc->u.low),
+		&instance->reg_set->inbound_low_queue_port);
+	writel(le32_to_cpu(req_desc->u.high),
+		&instance->reg_set->inbound_high_queue_port);
+	mmiowb();
+	spin_unlock_irqrestore(&instance->hba_lock, flags);
 #endif
-	}
 }
 
 /**
@@ -982,7 +976,6 @@
 	const char *sys_info;
 	MFI_CAPABILITIES *drv_ops;
 	u32 scratch_pad_2;
-	unsigned long flags;
 	ktime_t time;
 	bool cur_fw_64bit_dma_capable;
 
@@ -1121,14 +1114,7 @@
 			break;
 	}
 
-	/* For Ventura also IOC INIT required 64 bit Descriptor write. */
-	spin_lock_irqsave(&instance->hba_lock, flags);
-	writel(le32_to_cpu(req_desc.u.low),
-	       &instance->reg_set->inbound_low_queue_port);
-	writel(le32_to_cpu(req_desc.u.high),
-	       &instance->reg_set->inbound_high_queue_port);
-	mmiowb();
-	spin_unlock_irqrestore(&instance->hba_lock, flags);
+	megasas_fire_cmd_fusion(instance, &req_desc);
 
 	wait_and_poll(instance, cmd, MFI_POLL_TIMEOUT_SECS);
 
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c b/drivers/scsi/mpt3sas/mpt3sas_base.c
index 59a87ca..0aafbfd 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_base.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_base.c
@@ -6297,14 +6297,14 @@
 }
 
 /**
- * _wait_for_commands_to_complete - reset controller
+ * mpt3sas_wait_for_commands_to_complete - reset controller
  * @ioc: Pointer to MPT_ADAPTER structure
  *
  * This function is waiting 10s for all pending commands to complete
  * prior to putting controller in reset.
  */
-static void
-_wait_for_commands_to_complete(struct MPT3SAS_ADAPTER *ioc)
+void
+mpt3sas_wait_for_commands_to_complete(struct MPT3SAS_ADAPTER *ioc)
 {
 	u32 ioc_state;
 
@@ -6377,7 +6377,7 @@
 			is_fault = 1;
 	}
 	_base_reset_handler(ioc, MPT3_IOC_PRE_RESET);
-	_wait_for_commands_to_complete(ioc);
+	mpt3sas_wait_for_commands_to_complete(ioc);
 	_base_mask_interrupts(ioc);
 	r = _base_make_ioc_ready(ioc, type);
 	if (r)
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.h b/drivers/scsi/mpt3sas/mpt3sas_base.h
index 789bc42..99ccf83 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_base.h
+++ b/drivers/scsi/mpt3sas/mpt3sas_base.h
@@ -1433,6 +1433,9 @@
 
 int mpt3sas_port_enable(struct MPT3SAS_ADAPTER *ioc);
 
+void
+mpt3sas_wait_for_commands_to_complete(struct MPT3SAS_ADAPTER *ioc);
+
 
 /* scsih shared API */
 struct scsi_cmnd *mpt3sas_scsih_scsi_lookup_get(struct MPT3SAS_ADAPTER *ioc,
diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c
index 74fca18..c2ea13c7 100644
--- a/drivers/scsi/mpt3sas/mpt3sas_scsih.c
+++ b/drivers/scsi/mpt3sas/mpt3sas_scsih.c
@@ -2835,7 +2835,8 @@
 	_scsih_tm_display_info(ioc, scmd);
 
 	sas_device_priv_data = scmd->device->hostdata;
-	if (!sas_device_priv_data || !sas_device_priv_data->sas_target) {
+	if (!sas_device_priv_data || !sas_device_priv_data->sas_target ||
+	    ioc->remove_host) {
 		sdev_printk(KERN_INFO, scmd->device,
 			"device been deleted! scmd(%p)\n", scmd);
 		scmd->result = DID_NO_CONNECT << 16;
@@ -2898,7 +2899,8 @@
 	_scsih_tm_display_info(ioc, scmd);
 
 	sas_device_priv_data = scmd->device->hostdata;
-	if (!sas_device_priv_data || !sas_device_priv_data->sas_target) {
+	if (!sas_device_priv_data || !sas_device_priv_data->sas_target ||
+	    ioc->remove_host) {
 		sdev_printk(KERN_INFO, scmd->device,
 			"device been deleted! scmd(%p)\n", scmd);
 		scmd->result = DID_NO_CONNECT << 16;
@@ -2961,7 +2963,8 @@
 	_scsih_tm_display_info(ioc, scmd);
 
 	sas_device_priv_data = scmd->device->hostdata;
-	if (!sas_device_priv_data || !sas_device_priv_data->sas_target) {
+	if (!sas_device_priv_data || !sas_device_priv_data->sas_target ||
+	    ioc->remove_host) {
 		starget_printk(KERN_INFO, starget, "target been deleted! scmd(%p)\n",
 			scmd);
 		scmd->result = DID_NO_CONNECT << 16;
@@ -3019,7 +3022,7 @@
 	    ioc->name, scmd);
 	scsi_print_command(scmd);
 
-	if (ioc->is_driver_loading) {
+	if (ioc->is_driver_loading || ioc->remove_host) {
 		pr_info(MPT3SAS_FMT "Blocking the host reset\n",
 		    ioc->name);
 		r = FAILED;
@@ -4453,7 +4456,7 @@
 		st = scsi_cmd_priv(scmd);
 		mpt3sas_base_clear_st(ioc, st);
 		scsi_dma_unmap(scmd);
-		if (ioc->pci_error_recovery)
+		if (ioc->pci_error_recovery || ioc->remove_host)
 			scmd->result = DID_NO_CONNECT << 16;
 		else
 			scmd->result = DID_RESET << 16;
@@ -9739,6 +9742,10 @@
 	unsigned long flags;
 
 	ioc->remove_host = 1;
+
+	mpt3sas_wait_for_commands_to_complete(ioc);
+	_scsih_flush_running_cmds(ioc);
+
 	_scsih_fw_event_cleanup_queue(ioc);
 
 	spin_lock_irqsave(&ioc->fw_event_lock, flags);
@@ -9815,6 +9822,10 @@
 	unsigned long flags;
 
 	ioc->remove_host = 1;
+
+	mpt3sas_wait_for_commands_to_complete(ioc);
+	_scsih_flush_running_cmds(ioc);
+
 	_scsih_fw_event_cleanup_queue(ioc);
 
 	spin_lock_irqsave(&ioc->fw_event_lock, flags);
diff --git a/drivers/scsi/qedi/qedi_fw.c b/drivers/scsi/qedi/qedi_fw.c
index 667d769..d09afe1 100644
--- a/drivers/scsi/qedi/qedi_fw.c
+++ b/drivers/scsi/qedi/qedi_fw.c
@@ -762,6 +762,11 @@
 
 	iscsi_cid = cqe->conn_id;
 	qedi_conn = qedi->cid_que.conn_cid_tbl[iscsi_cid];
+	if (!qedi_conn) {
+		QEDI_INFO(&qedi->dbg_ctx, QEDI_LOG_INFO,
+			  "icid not found 0x%x\n", cqe->conn_id);
+		return;
+	}
 
 	/* Based on this itt get the corresponding qedi_cmd */
 	spin_lock_bh(&qedi_conn->tmf_work_lock);
diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h
index be7d682..c9689f9 100644
--- a/drivers/scsi/qla2xxx/qla_def.h
+++ b/drivers/scsi/qla2xxx/qla_def.h
@@ -261,9 +261,9 @@
 struct name_list_extended {
 	struct get_name_list_extended *l;
 	dma_addr_t		ldma;
-	struct list_head 	fcports;	/* protect by sess_list */
+	struct list_head	fcports;
+	spinlock_t		fcports_lock;
 	u32			size;
-	u8			sent;
 };
 /*
  * Timeout timer counts in seconds
@@ -2217,6 +2217,7 @@
 
 /* FCP-4 types */
 #define FC4_TYPE_FCP_SCSI	0x08
+#define FC4_TYPE_NVME		0x28
 #define FC4_TYPE_OTHER		0x0
 #define FC4_TYPE_UNKNOWN	0xff
 
diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
index 5bf9a59..403fa09 100644
--- a/drivers/scsi/qla2xxx/qla_gs.c
+++ b/drivers/scsi/qla2xxx/qla_gs.c
@@ -3179,6 +3179,7 @@
 	sp->free(sp);
 	fcport->flags &= ~FCF_ASYNC_SENT;
 done:
+	fcport->flags &= ~FCF_ASYNC_ACTIVE;
 	return rval;
 }
 
@@ -3370,6 +3371,7 @@
 	sp->free(sp);
 	fcport->flags &= ~FCF_ASYNC_SENT;
 done:
+	fcport->flags &= ~FCF_ASYNC_ACTIVE;
 	return rval;
 }
 
@@ -3971,6 +3973,9 @@
 	spin_lock_irqsave(&vha->work_lock, flags);
 	vha->scan.scan_flags &= ~SF_SCANNING;
 	spin_unlock_irqrestore(&vha->work_lock, flags);
+
+	if ((fc4type == FC4_TYPE_FCP_SCSI) && vha->flags.nvme_enabled)
+		qla24xx_async_gpnft(vha, FC4_TYPE_NVME);
 }
 
 static void qla2x00_async_gpnft_gnnft_sp_done(void *s, int res)
diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index 2dea112..00329dd 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -213,6 +213,7 @@
 	sp->free(sp);
 	fcport->flags &= ~FCF_ASYNC_SENT;
 done:
+	fcport->flags &= ~FCF_ASYNC_ACTIVE;
 	return rval;
 }
 
@@ -263,7 +264,7 @@
 done_free_sp:
 	sp->free(sp);
 done:
-	fcport->flags &= ~FCF_ASYNC_SENT;
+	fcport->flags &= ~(FCF_ASYNC_SENT | FCF_ASYNC_ACTIVE);
 	return rval;
 }
 
@@ -271,6 +272,7 @@
 qla2x00_async_prlo_done(struct scsi_qla_host *vha, fc_port_t *fcport,
     uint16_t *data)
 {
+	fcport->flags &= ~FCF_ASYNC_ACTIVE;
 	/* Don't re-login in target mode */
 	if (!fcport->tgt_session)
 		qla2x00_mark_device_lost(vha, fcport, 1, 0);
@@ -284,6 +286,7 @@
 	struct srb_iocb *lio = &sp->u.iocb_cmd;
 	struct scsi_qla_host *vha = sp->vha;
 
+	sp->fcport->flags &= ~FCF_ASYNC_ACTIVE;
 	if (!test_bit(UNLOADING, &vha->dpc_flags))
 		qla2x00_post_async_prlo_done_work(sp->fcport->vha, sp->fcport,
 		    lio->u.logio.data);
@@ -322,6 +325,7 @@
 done_free_sp:
 	sp->free(sp);
 done:
+	fcport->flags &= ~FCF_ASYNC_ACTIVE;
 	return rval;
 }
 
@@ -375,6 +379,8 @@
 	    "Async done-%s res %x %8phC\n",
 	    sp->name, res, sp->fcport->port_name);
 
+	sp->fcport->flags &= ~FCF_ASYNC_SENT;
+
 	memset(&ea, 0, sizeof(ea));
 	ea.event = FCME_ADISC_DONE;
 	ea.rc = res;
@@ -425,7 +431,7 @@
 done_free_sp:
 	sp->free(sp);
 done:
-	fcport->flags &= ~FCF_ASYNC_SENT;
+	fcport->flags &= ~(FCF_ASYNC_SENT | FCF_ASYNC_ACTIVE);
 	qla2x00_post_async_adisc_work(vha, fcport, data);
 	return rval;
 }
@@ -643,8 +649,7 @@
 		    (loop_id & 0x7fff));
 	}
 
-	spin_lock_irqsave(&vha->hw->tgt.sess_lock, flags);
-	vha->gnl.sent = 0;
+	spin_lock_irqsave(&vha->gnl.fcports_lock, flags);
 
 	INIT_LIST_HEAD(&h);
 	fcport = tf = NULL;
@@ -653,12 +658,16 @@
 
 	list_for_each_entry_safe(fcport, tf, &h, gnl_entry) {
 		list_del_init(&fcport->gnl_entry);
+		spin_lock(&vha->hw->tgt.sess_lock);
 		fcport->flags &= ~(FCF_ASYNC_SENT | FCF_ASYNC_ACTIVE);
+		spin_unlock(&vha->hw->tgt.sess_lock);
 		ea.fcport = fcport;
 
 		qla2x00_fcport_event_handler(vha, &ea);
 	}
+	spin_unlock_irqrestore(&vha->gnl.fcports_lock, flags);
 
+	spin_lock_irqsave(&vha->hw->tgt.sess_lock, flags);
 	/* create new fcport if fw has knowledge of new sessions */
 	for (i = 0; i < n; i++) {
 		port_id_t id;
@@ -710,18 +719,21 @@
 	ql_dbg(ql_dbg_disc, vha, 0x20d9,
 	    "Async-gnlist WWPN %8phC \n", fcport->port_name);
 
-	spin_lock_irqsave(&vha->hw->tgt.sess_lock, flags);
+	spin_lock_irqsave(&vha->gnl.fcports_lock, flags);
+	if (!list_empty(&fcport->gnl_entry)) {
+		spin_unlock_irqrestore(&vha->gnl.fcports_lock, flags);
+		rval = QLA_SUCCESS;
+		goto done;
+	}
+
+	spin_lock(&vha->hw->tgt.sess_lock);
 	fcport->disc_state = DSC_GNL;
 	fcport->last_rscn_gen = fcport->rscn_gen;
 	fcport->last_login_gen = fcport->login_gen;
+	spin_unlock(&vha->hw->tgt.sess_lock);
 
 	list_add_tail(&fcport->gnl_entry, &vha->gnl.fcports);
-	if (vha->gnl.sent) {
-		spin_unlock_irqrestore(&vha->hw->tgt.sess_lock, flags);
-		return QLA_SUCCESS;
-	}
-	vha->gnl.sent = 1;
-	spin_unlock_irqrestore(&vha->hw->tgt.sess_lock, flags);
+	spin_unlock_irqrestore(&vha->gnl.fcports_lock, flags);
 
 	sp = qla2x00_get_sp(vha, fcport, GFP_KERNEL);
 	if (!sp)
@@ -1049,6 +1061,7 @@
 	fc_port_t *fcport = ea->fcport;
 	struct port_database_24xx *pd;
 	struct srb *sp = ea->sp;
+	uint8_t	ls;
 
 	pd = (struct port_database_24xx *)sp->u.iocb_cmd.u.mbx.in;
 
@@ -1061,7 +1074,12 @@
 	if (fcport->disc_state == DSC_DELETE_PEND)
 		return;
 
-	switch (pd->current_login_state) {
+	if (fcport->fc4f_nvme)
+		ls = pd->current_login_state >> 4;
+	else
+		ls = pd->current_login_state & 0xf;
+
+	switch (ls) {
 	case PDS_PRLI_COMPLETE:
 		__qla24xx_parse_gpdb(vha, fcport, pd);
 		break;
@@ -1151,8 +1169,9 @@
 	if (fcport->scan_state != QLA_FCPORT_FOUND)
 		return 0;
 
-	if ((fcport->fw_login_state == DSC_LS_PLOGI_PEND) ||
-	    (fcport->fw_login_state == DSC_LS_PRLI_PEND))
+	if ((fcport->loop_id != FC_NO_LOOP_ID) &&
+	    ((fcport->fw_login_state == DSC_LS_PLOGI_PEND) ||
+	     (fcport->fw_login_state == DSC_LS_PRLI_PEND)))
 		return 0;
 
 	if (fcport->fw_login_state == DSC_LS_PLOGI_COMP) {
@@ -1527,6 +1546,7 @@
 	srb_t *sp = ptr;
 	struct srb_iocb *abt = &sp->u.iocb_cmd;
 
+	del_timer(&sp->u.iocb_cmd.timer);
 	complete(&abt->u.abt.comp);
 }
 
@@ -1791,6 +1811,7 @@
 	qla2x00_mark_device_lost(vha, fcport, 1, 0);
 	qlt_logo_completion_handler(fcport, data[0]);
 	fcport->login_gen++;
+	fcport->flags &= ~FCF_ASYNC_ACTIVE;
 	return;
 }
 
@@ -1798,6 +1819,7 @@
 qla2x00_async_adisc_done(struct scsi_qla_host *vha, fc_port_t *fcport,
     uint16_t *data)
 {
+	fcport->flags &= ~(FCF_ASYNC_SENT | FCF_ASYNC_ACTIVE);
 	if (data[0] == MBS_COMMAND_COMPLETE) {
 		qla2x00_update_fcport(vha, fcport);
 
@@ -1805,7 +1827,6 @@
 	}
 
 	/* Retry login. */
-	fcport->flags &= ~FCF_ASYNC_SENT;
 	if (data[1] & QLA_LOGIO_LOGIN_RETRIED)
 		set_bit(RELOGIN_NEEDED, &vha->dpc_flags);
 	else
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index afcb5567..285911e 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -4577,6 +4577,7 @@
 
 	spin_lock_init(&vha->work_lock);
 	spin_lock_init(&vha->cmd_list_lock);
+	spin_lock_init(&vha->gnl.fcports_lock);
 	init_waitqueue_head(&vha->fcport_waitQ);
 	init_waitqueue_head(&vha->vref_waitq);
 
@@ -4806,9 +4807,12 @@
 			fcport->d_id = e->u.new_sess.id;
 			fcport->flags |= FCF_FABRIC_DEVICE;
 			fcport->fw_login_state = DSC_LS_PLOGI_PEND;
-			if (e->u.new_sess.fc4_type == FC4_TYPE_FCP_SCSI)
+			if (e->u.new_sess.fc4_type == FC4_TYPE_FCP_SCSI) {
 				fcport->fc4_type = FC4_TYPE_FCP_SCSI;
-
+			} else if (e->u.new_sess.fc4_type == FC4_TYPE_NVME) {
+				fcport->fc4_type = FC4_TYPE_OTHER;
+				fcport->fc4f_nvme = FC4_TYPE_NVME;
+			}
 			memcpy(fcport->port_name, e->u.new_sess.port_name,
 			    WWN_SIZE);
 		} else {
@@ -4877,6 +4881,8 @@
 			}
 			qlt_plogi_ack_unref(vha, pla);
 		} else {
+			fc_port_t *dfcp = NULL;
+
 			spin_lock_irqsave(&vha->hw->tgt.sess_lock, flags);
 			tfcp = qla2x00_find_fcport_by_nportid(vha,
 			    &e->u.new_sess.id, 1);
@@ -4899,11 +4905,13 @@
 				default:
 					fcport->login_pause = 1;
 					tfcp->conflict = fcport;
-					qlt_schedule_sess_for_deletion(tfcp);
+					dfcp = tfcp;
 					break;
 				}
 			}
 			spin_unlock_irqrestore(&vha->hw->tgt.sess_lock, flags);
+			if (dfcp)
+				qlt_schedule_sess_for_deletion(tfcp);
 
 			wwn = wwn_to_u64(fcport->node_name);
 
diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c
index 896b2d8..b49ac85 100644
--- a/drivers/scsi/qla2xxx/qla_target.c
+++ b/drivers/scsi/qla2xxx/qla_target.c
@@ -1224,10 +1224,10 @@
 	}
 }
 
-/* ha->tgt.sess_lock supposed to be held on entry */
 void qlt_schedule_sess_for_deletion(struct fc_port *sess)
 {
 	struct qla_tgt *tgt = sess->tgt;
+	struct qla_hw_data *ha = sess->vha->hw;
 	unsigned long flags;
 
 	if (sess->disc_state == DSC_DELETE_PEND)
@@ -1244,16 +1244,16 @@
 			return;
 	}
 
+	spin_lock_irqsave(&ha->tgt.sess_lock, flags);
 	if (sess->deleted == QLA_SESS_DELETED)
 		sess->logout_on_delete = 0;
 
-	spin_lock_irqsave(&sess->vha->work_lock, flags);
 	if (sess->deleted == QLA_SESS_DELETION_IN_PROGRESS) {
-		spin_unlock_irqrestore(&sess->vha->work_lock, flags);
+		spin_unlock_irqrestore(&ha->tgt.sess_lock, flags);
 		return;
 	}
 	sess->deleted = QLA_SESS_DELETION_IN_PROGRESS;
-	spin_unlock_irqrestore(&sess->vha->work_lock, flags);
+	spin_unlock_irqrestore(&ha->tgt.sess_lock, flags);
 
 	sess->disc_state = DSC_DELETE_PEND;
 
@@ -1262,13 +1262,10 @@
 	ql_dbg(ql_dbg_tgt, sess->vha, 0xe001,
 	    "Scheduling sess %p for deletion\n", sess);
 
-	/* use cancel to push work element through before re-queue */
-	cancel_work_sync(&sess->del_work);
 	INIT_WORK(&sess->del_work, qla24xx_delete_sess_fn);
-	queue_work(sess->vha->hw->wq, &sess->del_work);
+	WARN_ON(!queue_work(sess->vha->hw->wq, &sess->del_work));
 }
 
-/* ha->tgt.sess_lock supposed to be held on entry */
 static void qlt_clear_tgt_db(struct qla_tgt *tgt)
 {
 	struct fc_port *sess;
@@ -1451,8 +1448,8 @@
 	ql_dbg(ql_dbg_tgt_mgt, vha, 0xf008, "qla_tgt_fc_port_deleted %p", sess);
 
 	sess->local = 1;
-	qlt_schedule_sess_for_deletion(sess);
 	spin_unlock_irqrestore(&vha->hw->tgt.sess_lock, flags);
+	qlt_schedule_sess_for_deletion(sess);
 }
 
 static inline int test_tgt_sess_count(struct qla_tgt *tgt)
@@ -1512,10 +1509,8 @@
 	 * Lock is needed, because we still can get an incoming packet.
 	 */
 	mutex_lock(&vha->vha_tgt.tgt_mutex);
-	spin_lock_irqsave(&ha->tgt.sess_lock, flags);
 	tgt->tgt_stop = 1;
 	qlt_clear_tgt_db(tgt);
-	spin_unlock_irqrestore(&ha->tgt.sess_lock, flags);
 	mutex_unlock(&vha->vha_tgt.tgt_mutex);
 	mutex_unlock(&qla_tgt_mutex);
 
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index d042915..ca53a5f 100644
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -223,7 +223,8 @@
 
 static void scsi_eh_inc_host_failed(struct rcu_head *head)
 {
-	struct Scsi_Host *shost = container_of(head, typeof(*shost), rcu);
+	struct scsi_cmnd *scmd = container_of(head, typeof(*scmd), rcu);
+	struct Scsi_Host *shost = scmd->device->host;
 	unsigned long flags;
 
 	spin_lock_irqsave(shost->host_lock, flags);
@@ -259,7 +260,7 @@
 	 * Ensure that all tasks observe the host state change before the
 	 * host_failed change.
 	 */
-	call_rcu(&shost->rcu, scsi_eh_inc_host_failed);
+	call_rcu(&scmd->rcu, scsi_eh_inc_host_failed);
 }
 
 /**
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index a86df9c..c84f931 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -671,6 +671,7 @@
 	if (!blk_rq_is_scsi(req)) {
 		WARN_ON_ONCE(!(cmd->flags & SCMD_INITIALIZED));
 		cmd->flags &= ~SCMD_INITIALIZED;
+		destroy_rcu_head(&cmd->rcu);
 	}
 
 	if (req->mq_ctx) {
@@ -720,6 +721,8 @@
 		int result)
 {
 	switch (host_byte(result)) {
+	case DID_OK:
+		return BLK_STS_OK;
 	case DID_TRANSPORT_FAILFAST:
 		return BLK_STS_TRANSPORT;
 	case DID_TARGET_FAILURE:
@@ -1151,6 +1154,7 @@
 	struct scsi_cmnd *cmd = blk_mq_rq_to_pdu(rq);
 
 	scsi_req_init(&cmd->req);
+	init_rcu_head(&cmd->rcu);
 	cmd->jiffies_at_alloc = jiffies;
 	cmd->retries = 0;
 }
diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c
index 6be5ab3..8c51d62 100644
--- a/drivers/scsi/storvsc_drv.c
+++ b/drivers/scsi/storvsc_drv.c
@@ -1311,7 +1311,8 @@
 			 */
 			cpumask_and(&alloced_mask, &stor_device->alloced_cpus,
 				    cpumask_of_node(cpu_to_node(q_num)));
-			for_each_cpu(tgt_cpu, &alloced_mask) {
+			for_each_cpu_wrap(tgt_cpu, &alloced_mask,
+					outgoing_channel->target_cpu + 1) {
 				if (tgt_cpu != outgoing_channel->target_cpu) {
 					outgoing_channel =
 					stor_device->stor_chns[tgt_cpu];
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index d8d4a90..2280b23 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -68,6 +68,9 @@
 	struct list_head list;  /* scsi_cmnd participates in queue lists */
 	struct list_head eh_entry; /* entry for the host eh_cmd_q */
 	struct delayed_work abort_work;
+
+	struct rcu_head rcu;
+
 	int eh_eflags;		/* Used by error handlr */
 
 	/*
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index 1a1df0d..a8b7bf8 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -571,8 +571,6 @@
 		struct blk_mq_tag_set	tag_set;
 	};
 
-	struct rcu_head rcu;
-
 	atomic_t host_busy;		   /* commands actually active on low-level */
 	atomic_t host_blocked;