Merge tag 'isci-for-3.5' into misc

isci update for 3.5

1/ Rework remote-node-context (RNC) handling for proper management of
   the silicon state machine in error handling and hot-plug conditions.
   Further details below, suffice to say if the RNC is mismanaged the
   silicon state machines may lock up.

2/ Refactor the initialization code to be reused for suspend/resume support

3/ Miscellaneous bug fixes to address discovery issues and hardware
   compatibility.

RNC rework details from Jeff Skirvin:

In the controller, devices as they appear on a SAS domain (or
direct-attached SATA devices) are represented by memory structures known
as "Remote Node Contexts" (RNCs).  These structures are transferred from
main memory to the controller using a set of register commands; these
commands include setting up the context ("posting"), removing the
context ("invalidating"), and commands to control the scheduling of
commands and connections to that remote device ("suspensions" and
"resumptions").  There is a similar path to control RNC scheduling from
the protocol engine, which interprets the results of command and data
transmission and reception.

In general, the controller chooses among non-suspended RNCs to find one
that has work requiring scheduling the transmission of command and data
frames to a target.  Likewise, when a target tries to return data back
to the initiator, the state of the RNC is used by the controller to
determine how to treat the incoming request. As an example, if the RNC
is in the state "TX/RX Suspended", incoming SSP connection requests from
the target will be rejected by the controller hardware.  When an RNC is
"TX Suspended", it will not be selected by the controller hardware to
start outgoing command or data operations (with certain priority-based
exceptions).

As mentioned above, there are two sources for management of the RNC
states: commands from driver software, and the result of transmission
and reception conditions of commands and data signaled by the controller
hardware.  As an example of the latter, if an outgoing SSP command ends
with a OPEN_REJECT(BAD_DESTINATION) status, the RNC state will
transition to the "TX Suspended" state, and this is signaled by the
controller hardware in the status to the completion of the pending
command as well as signaled in a controller hardware event.  Examples of
the former are included in the patch changelogs.

Driver software is required to suspend the RNC in a "TX/RX Suspended"
condition before any outstanding commands can be terminated.  Failure to
guarantee this can lead to a complete hardware hang condition.  Earlier
versions of the driver software did not guarantee that an RNC was
correctly managed before I/O termination, and so operated in an unsafe
way.

Further, the driver performed unnecessary contortions to preserve the
remote device command state and so was more complicated than it needed
to be.  A simplifying driver assumption is that once an I/O has entered
the error handler path without having completed in the target, the
requirement on the driver is that all use of the sas_task must end.
Beyond that, recovery of operation is dependent on libsas and other
components to reset, rediscover and reconfigure the device before normal
operation can restart.  In the driver, this simplifying assumption meant
that the RNC management could be reduced to entry into the suspended
state, terminating the targeted I/O request, and resuming the RNC as
needed for device-specific management such as an SSP Abort Task or LUN
Reset Management request.

1 files changed, 143 insertions, 657 deletions

diff --git a/drivers/scsi/isci/task.c b/drivers/scsi/isci/task.c
index 374254ede9d4..6bc74eb012c9 100644
--- a/drivers/scsi/isci/task.c
+++ b/drivers/scsi/isci/task.c
@@ -78,54 +78,25 @@ static void isci_task_refuse(struct isci_host *ihost, struct sas_task *task,
 			     enum exec_status status)
 
 {
-	enum isci_completion_selection disposition;
+	unsigned long flags;
 
-	disposition = isci_perform_normal_io_completion;
-	disposition = isci_task_set_completion_status(task, response, status,
-						      disposition);
+	/* Normal notification (task_done) */
+	dev_dbg(&ihost->pdev->dev, "%s: task = %p, response=%d, status=%d\n",
+		__func__, task, response, status);
 
-	/* Tasks aborted specifically by a call to the lldd_abort_task
-	 * function should not be completed to the host in the regular path.
-	 */
-	switch (disposition) {
-	case isci_perform_normal_io_completion:
-		/* Normal notification (task_done) */
-		dev_dbg(&ihost->pdev->dev,
-			"%s: Normal - task = %p, response=%d, "
-			"status=%d\n",
-			__func__, task, response, status);
-
-		task->lldd_task = NULL;
-		task->task_done(task);
-		break;
-
-	case isci_perform_aborted_io_completion:
-		/*
-		 * No notification because this request is already in the
-		 * abort path.
-		 */
-		dev_dbg(&ihost->pdev->dev,
-			"%s: Aborted - task = %p, response=%d, "
-			"status=%d\n",
-			__func__, task, response, status);
-		break;
+	spin_lock_irqsave(&task->task_state_lock, flags);
 
-	case isci_perform_error_io_completion:
-		/* Use sas_task_abort */
-		dev_dbg(&ihost->pdev->dev,
-			"%s: Error - task = %p, response=%d, "
-			"status=%d\n",
-			__func__, task, response, status);
-		sas_task_abort(task);
-		break;
+	task->task_status.resp = response;
+	task->task_status.stat = status;
 
-	default:
-		dev_dbg(&ihost->pdev->dev,
-			"%s: isci task notification default case!",
-			__func__);
-		sas_task_abort(task);
-		break;
-	}
+	/* Normal notification (task_done) */
+	task->task_state_flags |= SAS_TASK_STATE_DONE;
+	task->task_state_flags &= ~(SAS_TASK_AT_INITIATOR |
+				    SAS_TASK_STATE_PENDING);
+	task->lldd_task = NULL;
+	spin_unlock_irqrestore(&task->task_state_lock, flags);
+
+	task->task_done(task);
 }
 
 #define for_each_sas_task(num, task) \
@@ -289,60 +260,6 @@ static struct isci_request *isci_task_request_build(struct isci_host *ihost,
 	return ireq;
 }
 
-/**
-* isci_request_mark_zombie() - This function must be called with scic_lock held.
-*/
-static void isci_request_mark_zombie(struct isci_host *ihost, struct isci_request *ireq)
-{
-	struct completion *tmf_completion = NULL;
-	struct completion *req_completion;
-
-	/* Set the request state to "dead". */
-	ireq->status = dead;
-
-	req_completion = ireq->io_request_completion;
-	ireq->io_request_completion = NULL;
-
-	if (test_bit(IREQ_TMF, &ireq->flags)) {
-		/* Break links with the TMF request. */
-		struct isci_tmf *tmf = isci_request_access_tmf(ireq);
-
-		/* In the case where a task request is dying,
-		 * the thread waiting on the complete will sit and
-		 * timeout unless we wake it now.  Since the TMF
-		 * has a default error status, complete it here
-		 * to wake the waiting thread.
-		 */
-		if (tmf) {
-			tmf_completion = tmf->complete;
-			tmf->complete = NULL;
-		}
-		ireq->ttype_ptr.tmf_task_ptr = NULL;
-		dev_dbg(&ihost->pdev->dev, "%s: tmf_code %d, managed tag %#x\n",
-			__func__, tmf->tmf_code, tmf->io_tag);
-	} else {
-		/* Break links with the sas_task - the callback is done
-		 * elsewhere.
-		 */
-		struct sas_task *task = isci_request_access_task(ireq);
-
-		if (task)
-			task->lldd_task = NULL;
-
-		ireq->ttype_ptr.io_task_ptr = NULL;
-	}
-
-	dev_warn(&ihost->pdev->dev, "task context unrecoverable (tag: %#x)\n",
-		 ireq->io_tag);
-
-	/* Don't force waiting threads to timeout. */
-	if (req_completion)
-		complete(req_completion);
-
-	if (tmf_completion != NULL)
-		complete(tmf_completion);
-}
-
 static int isci_task_execute_tmf(struct isci_host *ihost,
 				 struct isci_remote_device *idev,
 				 struct isci_tmf *tmf, unsigned long timeout_ms)
@@ -400,17 +317,11 @@ static int isci_task_execute_tmf(struct isci_host *ihost,
 		spin_unlock_irqrestore(&ihost->scic_lock, flags);
 		goto err_tci;
 	}
-
-	if (tmf->cb_state_func != NULL)
-		tmf->cb_state_func(isci_tmf_started, tmf, tmf->cb_data);
-
-	isci_request_change_state(ireq, started);
-
-	/* add the request to the remote device request list. */
-	list_add(&ireq->dev_node, &idev->reqs_in_process);
-
 	spin_unlock_irqrestore(&ihost->scic_lock, flags);
 
+	/* The RNC must be unsuspended before the TMF can get a response. */
+	isci_remote_device_resume_from_abort(ihost, idev);
+
 	/* Wait for the TMF to complete, or a timeout. */
 	timeleft = wait_for_completion_timeout(&completion,
 					       msecs_to_jiffies(timeout_ms));
@@ -419,32 +330,7 @@ static int isci_task_execute_tmf(struct isci_host *ihost,
 		/* The TMF did not complete - this could be because
 		 * of an unplug.  Terminate the TMF request now.
 		 */
-		spin_lock_irqsave(&ihost->scic_lock, flags);
-
-		if (tmf->cb_state_func != NULL)
-			tmf->cb_state_func(isci_tmf_timed_out, tmf,
-					   tmf->cb_data);
-
-		sci_controller_terminate_request(ihost, idev, ireq);
-
-		spin_unlock_irqrestore(&ihost->scic_lock, flags);
-
-		timeleft = wait_for_completion_timeout(
-			&completion,
-			msecs_to_jiffies(ISCI_TERMINATION_TIMEOUT_MSEC));
-
-		if (!timeleft) {
-			/* Strange condition - the termination of the TMF
-			 * request timed-out.
-			 */
-			spin_lock_irqsave(&ihost->scic_lock, flags);
-
-			/* If the TMF status has not changed, kill it. */
-			if (tmf->status == SCI_FAILURE_TIMEOUT)
-				isci_request_mark_zombie(ihost, ireq);
-
-			spin_unlock_irqrestore(&ihost->scic_lock, flags);
-		}
+		isci_remote_device_suspend_terminate(ihost, idev, ireq);
 	}
 
 	isci_print_tmf(ihost, tmf);
@@ -476,315 +362,21 @@ static int isci_task_execute_tmf(struct isci_host *ihost,
 }
 
 static void isci_task_build_tmf(struct isci_tmf *tmf,
-				enum isci_tmf_function_codes code,
-				void (*tmf_sent_cb)(enum isci_tmf_cb_state,
-						    struct isci_tmf *,
-						    void *),
-				void *cb_data)
+				enum isci_tmf_function_codes code)
 {
 	memset(tmf, 0, sizeof(*tmf));
-
-	tmf->tmf_code      = code;
-	tmf->cb_state_func = tmf_sent_cb;
-	tmf->cb_data       = cb_data;
+	tmf->tmf_code = code;
 }
 
 static void isci_task_build_abort_task_tmf(struct isci_tmf *tmf,
 					   enum isci_tmf_function_codes code,
-					   void (*tmf_sent_cb)(enum isci_tmf_cb_state,
-							       struct isci_tmf *,
-							       void *),
 					   struct isci_request *old_request)
 {
-	isci_task_build_tmf(tmf, code, tmf_sent_cb, old_request);
+	isci_task_build_tmf(tmf, code);
 	tmf->io_tag = old_request->io_tag;
 }
 
 /**
- * isci_task_validate_request_to_abort() - This function checks the given I/O
- *    against the "started" state.  If the request is still "started", it's
- *    state is changed to aborted. NOTE: isci_host->scic_lock MUST BE HELD
- *    BEFORE CALLING THIS FUNCTION.
- * @isci_request: This parameter specifies the request object to control.
- * @isci_host: This parameter specifies the ISCI host object
- * @isci_device: This is the device to which the request is pending.
- * @aborted_io_completion: This is a completion structure that will be added to
- *    the request in case it is changed to aborting; this completion is
- *    triggered when the request is fully completed.
- *
- * Either "started" on successful change of the task status to "aborted", or
- * "unallocated" if the task cannot be controlled.
- */
-static enum isci_request_status isci_task_validate_request_to_abort(
-	struct isci_request *isci_request,
-	struct isci_host *isci_host,
-	struct isci_remote_device *isci_device,
-	struct completion *aborted_io_completion)
-{
-	enum isci_request_status old_state = unallocated;
-
-	/* Only abort the task if it's in the
-	 *  device's request_in_process list
-	 */
-	if (isci_request && !list_empty(&isci_request->dev_node)) {
-		old_state = isci_request_change_started_to_aborted(
-			isci_request, aborted_io_completion);
-
-	}
-
-	return old_state;
-}
-
-static int isci_request_is_dealloc_managed(enum isci_request_status stat)
-{
-	switch (stat) {
-	case aborted:
-	case aborting:
-	case terminating:
-	case completed:
-	case dead:
-		return true;
-	default:
-		return false;
-	}
-}
-
-/**
- * isci_terminate_request_core() - This function will terminate the given
- *    request, and wait for it to complete.  This function must only be called
- *    from a thread that can wait.  Note that the request is terminated and
- *    completed (back to the host, if started there).
- * @ihost: This SCU.
- * @idev: The target.
- * @isci_request: The I/O request to be terminated.
- *
- */
-static void isci_terminate_request_core(struct isci_host *ihost,
-					struct isci_remote_device *idev,
-					struct isci_request *isci_request)
-{
-	enum sci_status status      = SCI_SUCCESS;
-	bool was_terminated         = false;
-	bool needs_cleanup_handling = false;
-	unsigned long     flags;
-	unsigned long     termination_completed = 1;
-	struct completion *io_request_completion;
-
-	dev_dbg(&ihost->pdev->dev,
-		"%s: device = %p; request = %p\n",
-		__func__, idev, isci_request);
-
-	spin_lock_irqsave(&ihost->scic_lock, flags);
-
-	io_request_completion = isci_request->io_request_completion;
-
-	/* Note that we are not going to control
-	 * the target to abort the request.
-	 */
-	set_bit(IREQ_COMPLETE_IN_TARGET, &isci_request->flags);
-
-	/* Make sure the request wasn't just sitting around signalling
-	 * device condition (if the request handle is NULL, then the
-	 * request completed but needed additional handling here).
-	 */
-	if (!test_bit(IREQ_TERMINATED, &isci_request->flags)) {
-		was_terminated = true;
-		needs_cleanup_handling = true;
-		status = sci_controller_terminate_request(ihost,
-							   idev,
-							   isci_request);
-	}
-	spin_unlock_irqrestore(&ihost->scic_lock, flags);
-
-	/*
-	 * The only time the request to terminate will
-	 * fail is when the io request is completed and
-	 * being aborted.
-	 */
-	if (status != SCI_SUCCESS) {
-		dev_dbg(&ihost->pdev->dev,
-			"%s: sci_controller_terminate_request"
-			" returned = 0x%x\n",
-			__func__, status);
-
-		isci_request->io_request_completion = NULL;
-
-	} else {
-		if (was_terminated) {
-			dev_dbg(&ihost->pdev->dev,
-				"%s: before completion wait (%p/%p)\n",
-				__func__, isci_request, io_request_completion);
-
-			/* Wait here for the request to complete. */
-			termination_completed
-				= wait_for_completion_timeout(
-				   io_request_completion,
-				   msecs_to_jiffies(ISCI_TERMINATION_TIMEOUT_MSEC));
-
-			if (!termination_completed) {
-
-				/* The request to terminate has timed out.  */
-				spin_lock_irqsave(&ihost->scic_lock, flags);
-
-				/* Check for state changes. */
-				if (!test_bit(IREQ_TERMINATED,
-					      &isci_request->flags)) {
-
-					/* The best we can do is to have the
-					 * request die a silent death if it
-					 * ever really completes.
-					 */
-					isci_request_mark_zombie(ihost,
-								 isci_request);
-					needs_cleanup_handling = true;
-				} else
-					termination_completed = 1;
-
-				spin_unlock_irqrestore(&ihost->scic_lock,
-						       flags);
-
-				if (!termination_completed) {
-
-					dev_dbg(&ihost->pdev->dev,
-						"%s: *** Timeout waiting for "
-						"termination(%p/%p)\n",
-						__func__, io_request_completion,
-						isci_request);
-
-					/* The request can no longer be referenced
-					 * safely since it may go away if the
-					 * termination every really does complete.
-					 */
-					isci_request = NULL;
-				}
-			}
-			if (termination_completed)
-				dev_dbg(&ihost->pdev->dev,
-					"%s: after completion wait (%p/%p)\n",
-					__func__, isci_request, io_request_completion);
-		}
-
-		if (termination_completed) {
-
-			isci_request->io_request_completion = NULL;
-
-			/* Peek at the status of the request.  This will tell
-			 * us if there was special handling on the request such that it
-			 * needs to be detached and freed here.
-			 */
-			spin_lock_irqsave(&isci_request->state_lock, flags);
-
-			needs_cleanup_handling
-				= isci_request_is_dealloc_managed(
-					isci_request->status);
-
-			spin_unlock_irqrestore(&isci_request->state_lock, flags);
-
-		}
-		if (needs_cleanup_handling) {
-
-			dev_dbg(&ihost->pdev->dev,
-				"%s: cleanup isci_device=%p, request=%p\n",
-				__func__, idev, isci_request);
-
-			if (isci_request != NULL) {
-				spin_lock_irqsave(&ihost->scic_lock, flags);
-				isci_free_tag(ihost, isci_request->io_tag);
-				isci_request_change_state(isci_request, unallocated);
-				list_del_init(&isci_request->dev_node);
-				spin_unlock_irqrestore(&ihost->scic_lock, flags);
-			}
-		}
-	}
-}
-
-/**
- * isci_terminate_pending_requests() - This function will change the all of the
- *    requests on the given device's state to "aborting", will terminate the
- *    requests, and wait for them to complete.  This function must only be
- *    called from a thread that can wait.  Note that the requests are all
- *    terminated and completed (back to the host, if started there).
- * @isci_host: This parameter specifies SCU.
- * @idev: This parameter specifies the target.
- *
- */
-void isci_terminate_pending_requests(struct isci_host *ihost,
-				     struct isci_remote_device *idev)
-{
-	struct completion request_completion;
-	enum isci_request_status old_state;
-	unsigned long flags;
-	LIST_HEAD(list);
-
-	spin_lock_irqsave(&ihost->scic_lock, flags);
-	list_splice_init(&idev->reqs_in_process, &list);
-
-	/* assumes that isci_terminate_request_core deletes from the list */
-	while (!list_empty(&list)) {
-		struct isci_request *ireq = list_entry(list.next, typeof(*ireq), dev_node);
-
-		/* Change state to "terminating" if it is currently
-		 * "started".
-		 */
-		old_state = isci_request_change_started_to_newstate(ireq,
-								    &request_completion,
-								    terminating);
-		switch (old_state) {
-		case started:
-		case completed:
-		case aborting:
-			break;
-		default:
-			/* termination in progress, or otherwise dispositioned.
-			 * We know the request was on 'list' so should be safe
-			 * to move it back to reqs_in_process
-			 */
-			list_move(&ireq->dev_node, &idev->reqs_in_process);
-			ireq = NULL;
-			break;
-		}
-
-		if (!ireq)
-			continue;
-		spin_unlock_irqrestore(&ihost->scic_lock, flags);
-
-		init_completion(&request_completion);
-
-		dev_dbg(&ihost->pdev->dev,
-			 "%s: idev=%p request=%p; task=%p old_state=%d\n",
-			 __func__, idev, ireq,
-			(!test_bit(IREQ_TMF, &ireq->flags)
-				? isci_request_access_task(ireq)
-				: NULL),
-			old_state);
-
-		/* If the old_state is started:
-		 * This request was not already being aborted. If it had been,
-		 * then the aborting I/O (ie. the TMF request) would not be in
-		 * the aborting state, and thus would be terminated here.  Note
-		 * that since the TMF completion's call to the kernel function
-		 * "complete()" does not happen until the pending I/O request
-		 * terminate fully completes, we do not have to implement a
-		 * special wait here for already aborting requests - the
-		 * termination of the TMF request will force the request
-		 * to finish it's already started terminate.
-		 *
-		 * If old_state == completed:
-		 * This request completed from the SCU hardware perspective
-		 * and now just needs cleaning up in terms of freeing the
-		 * request and potentially calling up to libsas.
-		 *
-		 * If old_state == aborting:
-		 * This request has already gone through a TMF timeout, but may
-		 * not have been terminated; needs cleaning up at least.
-		 */
-		isci_terminate_request_core(ihost, idev, ireq);
-		spin_lock_irqsave(&ihost->scic_lock, flags);
-	}
-	spin_unlock_irqrestore(&ihost->scic_lock, flags);
-}
-
-/**
  * isci_task_send_lu_reset_sas() - This function is called by of the SAS Domain
  *    Template functions.
  * @lun: This parameter specifies the lun to be reset.
@@ -807,7 +399,7 @@ static int isci_task_send_lu_reset_sas(
 	 * value is "TMF_RESP_FUNC_COMPLETE", or the request timed-out (or
 	 * was otherwise unable to be executed ("TMF_RESP_FUNC_FAILED").
 	 */
-	isci_task_build_tmf(&tmf, isci_tmf_ssp_lun_reset, NULL, NULL);
+	isci_task_build_tmf(&tmf, isci_tmf_ssp_lun_reset);
 
 	#define ISCI_LU_RESET_TIMEOUT_MS 2000 /* 2 second timeout. */
 	ret = isci_task_execute_tmf(isci_host, isci_device, &tmf, ISCI_LU_RESET_TIMEOUT_MS);
@@ -826,42 +418,44 @@ static int isci_task_send_lu_reset_sas(
 
 int isci_task_lu_reset(struct domain_device *dev, u8 *lun)
 {
-	struct isci_host *isci_host = dev_to_ihost(dev);
-	struct isci_remote_device *isci_device;
+	struct isci_host *ihost = dev_to_ihost(dev);
+	struct isci_remote_device *idev;
 	unsigned long flags;
-	int ret;
+	int ret = TMF_RESP_FUNC_COMPLETE;
 
-	spin_lock_irqsave(&isci_host->scic_lock, flags);
-	isci_device = isci_lookup_device(dev);
-	spin_unlock_irqrestore(&isci_host->scic_lock, flags);
+	spin_lock_irqsave(&ihost->scic_lock, flags);
+	idev = isci_get_device(dev->lldd_dev);
+	spin_unlock_irqrestore(&ihost->scic_lock, flags);
 
-	dev_dbg(&isci_host->pdev->dev,
+	dev_dbg(&ihost->pdev->dev,
 		"%s: domain_device=%p, isci_host=%p; isci_device=%p\n",
-		 __func__, dev, isci_host, isci_device);
+		__func__, dev, ihost, idev);
 
-	if (!isci_device) {
-		/* If the device is gone, stop the escalations. */
-		dev_dbg(&isci_host->pdev->dev, "%s: No dev\n", __func__);
+	if (!idev) {
+		/* If the device is gone, escalate to I_T_Nexus_Reset. */
+		dev_dbg(&ihost->pdev->dev, "%s: No dev\n", __func__);
 
-		ret = TMF_RESP_FUNC_COMPLETE;
+		ret = TMF_RESP_FUNC_FAILED;
 		goto out;
 	}
 
-	/* Send the task management part of the reset. */
-	if (dev_is_sata(dev)) {
-		sas_ata_schedule_reset(dev);
-		ret = TMF_RESP_FUNC_COMPLETE;
-	} else
-		ret = isci_task_send_lu_reset_sas(isci_host, isci_device, lun);
-
-	/* If the LUN reset worked, all the I/O can now be terminated. */
-	if (ret == TMF_RESP_FUNC_COMPLETE)
-		/* Terminate all I/O now. */
-		isci_terminate_pending_requests(isci_host,
-						isci_device);
-
+	/* Suspend the RNC, kill all TCs */
+	if (isci_remote_device_suspend_terminate(ihost, idev, NULL)
+	    != SCI_SUCCESS) {
+		/* The suspend/terminate only fails if isci_get_device fails */
+		ret = TMF_RESP_FUNC_FAILED;
+		goto out;
+	}
+	/* All pending I/Os have been terminated and cleaned up. */
+	if (!test_bit(IDEV_GONE, &idev->flags)) {
+		if (dev_is_sata(dev))
+			sas_ata_schedule_reset(dev);
+		else
+			/* Send the task management part of the reset. */
+			ret = isci_task_send_lu_reset_sas(ihost, idev, lun);
+	}
  out:
-	isci_put_device(isci_device);
+	isci_put_device(idev);
 	return ret;
 }
 
@@ -882,63 +476,6 @@ int isci_task_clear_nexus_ha(struct sas_ha_struct *ha)
 /* Task Management Functions. Must be called from process context.	 */
 
 /**
- * isci_abort_task_process_cb() - This is a helper function for the abort task
- *    TMF command.  It manages the request state with respect to the successful
- *    transmission / completion of the abort task request.
- * @cb_state: This parameter specifies when this function was called - after
- *    the TMF request has been started and after it has timed-out.
- * @tmf: This parameter specifies the TMF in progress.
- *
- *
- */
-static void isci_abort_task_process_cb(
-	enum isci_tmf_cb_state cb_state,
-	struct isci_tmf *tmf,
-	void *cb_data)
-{
-	struct isci_request *old_request;
-
-	old_request = (struct isci_request *)cb_data;
-
-	dev_dbg(&old_request->isci_host->pdev->dev,
-		"%s: tmf=%p, old_request=%p\n",
-		__func__, tmf, old_request);
-
-	switch (cb_state) {
-
-	case isci_tmf_started:
-		/* The TMF has been started.  Nothing to do here, since the
-		 * request state was already set to "aborted" by the abort
-		 * task function.
-		 */
-		if ((old_request->status != aborted)
-			&& (old_request->status != completed))
-			dev_dbg(&old_request->isci_host->pdev->dev,
-				"%s: Bad request status (%d): tmf=%p, old_request=%p\n",
-				__func__, old_request->status, tmf, old_request);
-		break;
-
-	case isci_tmf_timed_out:
-
-		/* Set the task's state to "aborting", since the abort task
-		 * function thread set it to "aborted" (above) in anticipation
-		 * of the task management request working correctly.  Since the
-		 * timeout has now fired, the TMF request failed.  We set the
-		 * state such that the request completion will indicate the
-		 * device is no longer present.
-		 */
-		isci_request_change_state(old_request, aborting);
-		break;
-
-	default:
-		dev_dbg(&old_request->isci_host->pdev->dev,
-			"%s: Bad cb_state (%d): tmf=%p, old_request=%p\n",
-			__func__, cb_state, tmf, old_request);
-		break;
-	}
-}
-
-/**
  * isci_task_abort_task() - This function is one of the SAS Domain Template
  *    functions. This function is called by libsas to abort a specified task.
  * @task: This parameter specifies the SAS task to abort.
@@ -947,22 +484,20 @@ static void isci_abort_task_process_cb(
  */
 int isci_task_abort_task(struct sas_task *task)
 {
-	struct isci_host *isci_host = dev_to_ihost(task->dev);
+	struct isci_host *ihost = dev_to_ihost(task->dev);
 	DECLARE_COMPLETION_ONSTACK(aborted_io_completion);
 	struct isci_request       *old_request = NULL;
-	enum isci_request_status  old_state;
-	struct isci_remote_device *isci_device = NULL;
+	struct isci_remote_device *idev = NULL;
 	struct isci_tmf           tmf;
 	int                       ret = TMF_RESP_FUNC_FAILED;
 	unsigned long             flags;
-	int                       perform_termination = 0;
 
 	/* Get the isci_request reference from the task.  Note that
 	 * this check does not depend on the pending request list
 	 * in the device, because tasks driving resets may land here
 	 * after completion in the core.
 	 */
-	spin_lock_irqsave(&isci_host->scic_lock, flags);
+	spin_lock_irqsave(&ihost->scic_lock, flags);
 	spin_lock(&task->task_state_lock);
 
 	old_request = task->lldd_task;
@@ -971,20 +506,29 @@ int isci_task_abort_task(struct sas_task *task)
 	if (!(task->task_state_flags & SAS_TASK_STATE_DONE) &&
 	    (task->task_state_flags & SAS_TASK_AT_INITIATOR) &&
 	    old_request)
-		isci_device = isci_lookup_device(task->dev);
+		idev = isci_get_device(task->dev->lldd_dev);
 
 	spin_unlock(&task->task_state_lock);
-	spin_unlock_irqrestore(&isci_host->scic_lock, flags);
+	spin_unlock_irqrestore(&ihost->scic_lock, flags);
 
-	dev_dbg(&isci_host->pdev->dev,
-		"%s: dev = %p, task = %p, old_request == %p\n",
-		__func__, isci_device, task, old_request);
+	dev_warn(&ihost->pdev->dev,
+		 "%s: dev = %p (%s%s), task = %p, old_request == %p\n",
+		 __func__, idev,
+		 (dev_is_sata(task->dev) ? "STP/SATA"
+					 : ((dev_is_expander(task->dev))
+						? "SMP"
+						: "SSP")),
+		 ((idev) ? ((test_bit(IDEV_GONE, &idev->flags))
+			   ? " IDEV_GONE"
+			   : "")
+			 : " <NULL>"),
+		 task, old_request);
 
 	/* Device reset conditions signalled in task_state_flags are the
 	 * responsbility of libsas to observe at the start of the error
 	 * handler thread.
 	 */
-	if (!isci_device || !old_request) {
+	if (!idev || !old_request) {
 		/* The request has already completed and there
 		* is nothing to do here other than to set the task
 		* done bit, and indicate that the task abort function
@@ -998,108 +542,72 @@ int isci_task_abort_task(struct sas_task *task)
 
 		ret = TMF_RESP_FUNC_COMPLETE;
 
-		dev_dbg(&isci_host->pdev->dev,
-			"%s: abort task not needed for %p\n",
-			__func__, task);
+		dev_warn(&ihost->pdev->dev,
+			 "%s: abort task not needed for %p\n",
+			 __func__, task);
 		goto out;
 	}
-
-	spin_lock_irqsave(&isci_host->scic_lock, flags);
-
-	/* Check the request status and change to "aborted" if currently
-	 * "starting"; if true then set the I/O kernel completion
-	 * struct that will be triggered when the request completes.
-	 */
-	old_state = isci_task_validate_request_to_abort(
-				old_request, isci_host, isci_device,
-				&aborted_io_completion);
-	if ((old_state != started) &&
-	    (old_state != completed) &&
-	    (old_state != aborting)) {
-
-		spin_unlock_irqrestore(&isci_host->scic_lock, flags);
-
-		/* The request was already being handled by someone else (because
-		* they got to set the state away from started).
-		*/
-		dev_dbg(&isci_host->pdev->dev,
-			"%s:  device = %p; old_request %p already being aborted\n",
-			__func__,
-			isci_device, old_request);
-		ret = TMF_RESP_FUNC_COMPLETE;
+	/* Suspend the RNC, kill the TC */
+	if (isci_remote_device_suspend_terminate(ihost, idev, old_request)
+	    != SCI_SUCCESS) {
+		dev_warn(&ihost->pdev->dev,
+			 "%s: isci_remote_device_reset_terminate(dev=%p, "
+				 "req=%p, task=%p) failed\n",
+			 __func__, idev, old_request, task);
+		ret = TMF_RESP_FUNC_FAILED;
 		goto out;
 	}
+	spin_lock_irqsave(&ihost->scic_lock, flags);
+
 	if (task->task_proto == SAS_PROTOCOL_SMP ||
 	    sas_protocol_ata(task->task_proto) ||
-	    test_bit(IREQ_COMPLETE_IN_TARGET, &old_request->flags)) {
+	    test_bit(IREQ_COMPLETE_IN_TARGET, &old_request->flags) ||
+	    test_bit(IDEV_GONE, &idev->flags)) {
 
-		spin_unlock_irqrestore(&isci_host->scic_lock, flags);
+		spin_unlock_irqrestore(&ihost->scic_lock, flags);
 
-		dev_dbg(&isci_host->pdev->dev,
-			"%s: %s request"
-			" or complete_in_target (%d), thus no TMF\n",
-			__func__,
-			((task->task_proto == SAS_PROTOCOL_SMP)
-				? "SMP"
-				: (sas_protocol_ata(task->task_proto)
-					? "SATA/STP"
-					: "<other>")
-			 ),
-			test_bit(IREQ_COMPLETE_IN_TARGET, &old_request->flags));
-
-		if (test_bit(IREQ_COMPLETE_IN_TARGET, &old_request->flags)) {
-			spin_lock_irqsave(&task->task_state_lock, flags);
-			task->task_state_flags |= SAS_TASK_STATE_DONE;
-			task->task_state_flags &= ~(SAS_TASK_AT_INITIATOR |
-						    SAS_TASK_STATE_PENDING);
-			spin_unlock_irqrestore(&task->task_state_lock, flags);
-			ret = TMF_RESP_FUNC_COMPLETE;
-		} else {
-			spin_lock_irqsave(&task->task_state_lock, flags);
-			task->task_state_flags &= ~(SAS_TASK_AT_INITIATOR |
-						    SAS_TASK_STATE_PENDING);
-			spin_unlock_irqrestore(&task->task_state_lock, flags);
-		}
+		/* No task to send, so explicitly resume the device here */
+		isci_remote_device_resume_from_abort(ihost, idev);
 
-		/* STP and SMP devices are not sent a TMF, but the
-		 * outstanding I/O request is terminated below.  This is
-		 * because SATA/STP and SMP discovery path timeouts directly
-		 * call the abort task interface for cleanup.
-		 */
-		perform_termination = 1;
+		dev_warn(&ihost->pdev->dev,
+			 "%s: %s request"
+				 " or complete_in_target (%d), "
+				 "or IDEV_GONE (%d), thus no TMF\n",
+			 __func__,
+			 ((task->task_proto == SAS_PROTOCOL_SMP)
+			  ? "SMP"
+			  : (sas_protocol_ata(task->task_proto)
+				? "SATA/STP"
+				: "<other>")
+			  ),
+			 test_bit(IREQ_COMPLETE_IN_TARGET,
+				  &old_request->flags),
+			 test_bit(IDEV_GONE, &idev->flags));
+
+		spin_lock_irqsave(&task->task_state_lock, flags);
+		task->task_state_flags &= ~(SAS_TASK_AT_INITIATOR |
+					    SAS_TASK_STATE_PENDING);
+		task->task_state_flags |= SAS_TASK_STATE_DONE;
+		spin_unlock_irqrestore(&task->task_state_lock, flags);
 
+		ret = TMF_RESP_FUNC_COMPLETE;
 	} else {
 		/* Fill in the tmf stucture */
 		isci_task_build_abort_task_tmf(&tmf, isci_tmf_ssp_task_abort,
-					       isci_abort_task_process_cb,
 					       old_request);
 
-		spin_unlock_irqrestore(&isci_host->scic_lock, flags);
+		spin_unlock_irqrestore(&ihost->scic_lock, flags);
 
+		/* Send the task management request. */
 		#define ISCI_ABORT_TASK_TIMEOUT_MS 500 /* 1/2 second timeout */
-		ret = isci_task_execute_tmf(isci_host, isci_device, &tmf,
+		ret = isci_task_execute_tmf(ihost, idev, &tmf,
 					    ISCI_ABORT_TASK_TIMEOUT_MS);
-
-		if (ret == TMF_RESP_FUNC_COMPLETE)
-			perform_termination = 1;
-		else
-			dev_dbg(&isci_host->pdev->dev,
-				"%s: isci_task_send_tmf failed\n", __func__);
 	}
-	if (perform_termination) {
-		set_bit(IREQ_COMPLETE_IN_TARGET, &old_request->flags);
-
-		/* Clean up the request on our side, and wait for the aborted
-		 * I/O to complete.
-		 */
-		isci_terminate_request_core(isci_host, isci_device,
-					    old_request);
-	}
-
-	/* Make sure we do not leave a reference to aborted_io_completion */
-	old_request->io_request_completion = NULL;
- out:
-	isci_put_device(isci_device);
+out:
+	dev_warn(&ihost->pdev->dev,
+		 "%s: Done; dev = %p, task = %p , old_request == %p\n",
+		 __func__, idev, task, old_request);
+	isci_put_device(idev);
 	return ret;
 }
 
@@ -1195,14 +703,11 @@ isci_task_request_complete(struct isci_host *ihost,
 {
 	struct isci_tmf *tmf = isci_request_access_tmf(ireq);
 	struct completion *tmf_complete = NULL;
-	struct completion *request_complete = ireq->io_request_completion;
 
 	dev_dbg(&ihost->pdev->dev,
 		"%s: request = %p, status=%d\n",
 		__func__, ireq, completion_status);
 
-	isci_request_change_state(ireq, completed);
-
 	set_bit(IREQ_COMPLETE_IN_TARGET, &ireq->flags);
 
 	if (tmf) {
@@ -1226,20 +731,11 @@ isci_task_request_complete(struct isci_host *ihost,
 	 */
 	set_bit(IREQ_TERMINATED, &ireq->flags);
 
-	/* As soon as something is in the terminate path, deallocation is
-	 * managed there.  Note that the final non-managed state of a task
-	 * request is "completed".
-	 */
-	if ((ireq->status == completed) ||
-	    !isci_request_is_dealloc_managed(ireq->status)) {
-		isci_request_change_state(ireq, unallocated);
-		isci_free_tag(ihost, ireq->io_tag);
-		list_del_init(&ireq->dev_node);
-	}
+	if (test_and_clear_bit(IREQ_ABORT_PATH_ACTIVE, &ireq->flags))
+		wake_up_all(&ihost->eventq);
 
-	/* "request_complete" is set if the task was being terminated. */
-	if (request_complete)
-		complete(request_complete);
+	if (!test_bit(IREQ_NO_AUTO_FREE_TAG, &ireq->flags))
+		isci_free_tag(ihost, ireq->io_tag);
 
 	/* The task management part completes last. */
 	if (tmf_complete)
@@ -1250,48 +746,38 @@ static int isci_reset_device(struct isci_host *ihost,
 			     struct domain_device *dev,
 			     struct isci_remote_device *idev)
 {
-	int rc;
-	unsigned long flags;
-	enum sci_status status;
+	int rc = TMF_RESP_FUNC_COMPLETE, reset_stat = -1;
 	struct sas_phy *phy = sas_get_local_phy(dev);
 	struct isci_port *iport = dev->port->lldd_port;
 
 	dev_dbg(&ihost->pdev->dev, "%s: idev %p\n", __func__, idev);
 
-	spin_lock_irqsave(&ihost->scic_lock, flags);
-	status = sci_remote_device_reset(idev);
-	spin_unlock_irqrestore(&ihost->scic_lock, flags);
-
-	if (status != SCI_SUCCESS) {
-		dev_dbg(&ihost->pdev->dev,
-			 "%s: sci_remote_device_reset(%p) returned %d!\n",
-			 __func__, idev, status);
+	/* Suspend the RNC, terminate all outstanding TCs. */
+	if (isci_remote_device_suspend_terminate(ihost, idev, NULL)
+	    != SCI_SUCCESS) {
 		rc = TMF_RESP_FUNC_FAILED;
 		goto out;
 	}
-
-	if (scsi_is_sas_phy_local(phy)) {
-		struct isci_phy *iphy = &ihost->phys[phy->number];
-
-		rc = isci_port_perform_hard_reset(ihost, iport, iphy);
-	} else
-		rc = sas_phy_reset(phy, !dev_is_sata(dev));
-
-	/* Terminate in-progress I/O now. */
-	isci_remote_device_nuke_requests(ihost, idev);
-
-	/* Since all pending TCs have been cleaned, resume the RNC. */
-	spin_lock_irqsave(&ihost->scic_lock, flags);
-	status = sci_remote_device_reset_complete(idev);
-	spin_unlock_irqrestore(&ihost->scic_lock, flags);
-
-	if (status != SCI_SUCCESS) {
-		dev_dbg(&ihost->pdev->dev,
-			 "%s: sci_remote_device_reset_complete(%p) "
-			 "returned %d!\n", __func__, idev, status);
+	/* Note that since the termination for outstanding requests succeeded,
+	 * this function will return success.  This is because the resets will
+	 * only fail if the device has been removed (ie. hotplug), and the
+	 * primary duty of this function is to cleanup tasks, so that is the
+	 * relevant status.
+	 */
+	if (!test_bit(IDEV_GONE, &idev->flags)) {
+		if (scsi_is_sas_phy_local(phy)) {
+			struct isci_phy *iphy = &ihost->phys[phy->number];
+
+			reset_stat = isci_port_perform_hard_reset(ihost, iport,
+								  iphy);
+		} else
+			reset_stat = sas_phy_reset(phy, !dev_is_sata(dev));
 	}
+	/* Explicitly resume the RNC here, since there was no task sent. */
+	isci_remote_device_resume_from_abort(ihost, idev);
 
-	dev_dbg(&ihost->pdev->dev, "%s: idev %p complete.\n", __func__, idev);
+	dev_dbg(&ihost->pdev->dev, "%s: idev %p complete, reset_stat=%d.\n",
+		__func__, idev, reset_stat);
  out:
 	sas_put_local_phy(phy);
 	return rc;
@@ -1305,7 +791,7 @@ int isci_task_I_T_nexus_reset(struct domain_device *dev)
 	int ret;
 
 	spin_lock_irqsave(&ihost->scic_lock, flags);
-	idev = isci_lookup_device(dev);
+	idev = isci_get_device(dev->lldd_dev);
 	spin_unlock_irqrestore(&ihost->scic_lock, flags);
 
 	if (!idev) {