cxgb4/iw_cxgb4: Doorbell Drop Avoidance Bug Fixes

The current logic suffers from a slow response time to disable user DB
usage, and also fails to avoid DB FIFO drops under heavy load. This commit
fixes these deficiencies and makes the avoidance logic more optimal.
This is done by more efficiently notifying the ULDs of potential DB
problems, and implements a smoother flow control algorithm in iw_cxgb4,
which is the ULD that puts the most load on the DB fifo.

Design:

cxgb4:

Direct ULD callback from the DB FULL/DROP interrupt handler.  This allows
the ULD to stop doing user DB writes as quickly as possible.

While user DB usage is disabled, the LLD will accumulate DB write events
for its queues.  Then once DB usage is reenabled, a single DB write is
done for each queue with its accumulated write count.  This reduces the
load put on the DB fifo when reenabling.

iw_cxgb4:

Instead of marking each qp to indicate DB writes are disabled, we create
a device-global status page that each user process maps.  This allows
iw_cxgb4 to only set this single bit to disable all DB writes for all
user QPs vs traversing the idr of all the active QPs.  If the libcxgb4
doesn't support this, then we fall back to the old approach of marking
each QP.  Thus we allow the new driver to work with an older libcxgb4.

When the LLD upcalls iw_cxgb4 indicating DB FULL, we disable all DB writes
via the status page and transition the DB state to STOPPED.  As user
processes see that DB writes are disabled, they call into iw_cxgb4
to submit their DB write events.  Since the DB state is in STOPPED,
the QP trying to write gets enqueued on a new DB "flow control" list.
As subsequent DB writes are submitted for this flow controlled QP, the
amount of writes are accumulated for each QP on the flow control list.
So all the user QPs that are actively ringing the DB get put on this
list and the number of writes they request are accumulated.

When the LLD upcalls iw_cxgb4 indicating DB EMPTY, which is in a workq
context, we change the DB state to FLOW_CONTROL, and begin resuming all
the QPs that are on the flow control list.  This logic runs on until
the flow control list is empty or we exit FLOW_CONTROL mode (due to
a DB DROP upcall, for example).  QPs are removed from this list, and
their accumulated DB write counts written to the DB FIFO.  Sets of QPs,
called chunks in the code, are removed at one time. The chunk size is 64.
So 64 QPs are resumed at a time, and before the next chunk is resumed, the
logic waits (blocks) for the DB FIFO to drain.  This prevents resuming to
quickly and overflowing the FIFO.  Once the flow control list is empty,
the db state transitions back to NORMAL and user QPs are again allowed
to write directly to the user DB register.

The algorithm is designed such that if the DB write load is high enough,
then all the DB writes get submitted by the kernel using this flow
controlled approach to avoid DB drops.  As the load lightens though, we
resume to normal DB writes directly by user applications.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

9 files changed, 286 insertions, 190 deletions

diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 4a033853312e..ba7335fd4ebf 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -64,6 +64,10 @@ struct uld_ctx {
 static LIST_HEAD(uld_ctx_list);
 static DEFINE_MUTEX(dev_mutex);
 
+#define DB_FC_RESUME_SIZE 64
+#define DB_FC_RESUME_DELAY 1
+#define DB_FC_DRAIN_THRESH 0
+
 static struct dentry *c4iw_debugfs_root;
 
 struct c4iw_debugfs_data {
@@ -282,7 +286,7 @@ static const struct file_operations stag_debugfs_fops = {
 	.llseek  = default_llseek,
 };
 
-static char *db_state_str[] = {"NORMAL", "FLOW_CONTROL", "RECOVERY"};
+static char *db_state_str[] = {"NORMAL", "FLOW_CONTROL", "RECOVERY", "STOPPED"};
 
 static int stats_show(struct seq_file *seq, void *v)
 {
@@ -311,9 +315,10 @@ static int stats_show(struct seq_file *seq, void *v)
 	seq_printf(seq, "  DB FULL: %10llu\n", dev->rdev.stats.db_full);
 	seq_printf(seq, " DB EMPTY: %10llu\n", dev->rdev.stats.db_empty);
 	seq_printf(seq, "  DB DROP: %10llu\n", dev->rdev.stats.db_drop);
-	seq_printf(seq, " DB State: %s Transitions %llu\n",
+	seq_printf(seq, " DB State: %s Transitions %llu FC Interruptions %llu\n",
 		   db_state_str[dev->db_state],
-		   dev->rdev.stats.db_state_transitions);
+		   dev->rdev.stats.db_state_transitions,
+		   dev->rdev.stats.db_fc_interruptions);
 	seq_printf(seq, "TCAM_FULL: %10llu\n", dev->rdev.stats.tcam_full);
 	seq_printf(seq, "ACT_OFLD_CONN_FAILS: %10llu\n",
 		   dev->rdev.stats.act_ofld_conn_fails);
@@ -643,6 +648,12 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
 		printk(KERN_ERR MOD "error %d initializing ocqp pool\n", err);
 		goto err4;
 	}
+	rdev->status_page = (struct t4_dev_status_page *)
+			    __get_free_page(GFP_KERNEL);
+	if (!rdev->status_page) {
+		pr_err(MOD "error allocating status page\n");
+		goto err4;
+	}
 	return 0;
 err4:
 	c4iw_rqtpool_destroy(rdev);
@@ -656,6 +667,7 @@ err1:
 
 static void c4iw_rdev_close(struct c4iw_rdev *rdev)
 {
+	free_page((unsigned long)rdev->status_page);
 	c4iw_pblpool_destroy(rdev);
 	c4iw_rqtpool_destroy(rdev);
 	c4iw_destroy_resource(&rdev->resource);
@@ -703,18 +715,6 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 		pr_info("%s: On-Chip Queues not supported on this device.\n",
 			pci_name(infop->pdev));
 
-	if (!is_t4(infop->adapter_type)) {
-		if (!allow_db_fc_on_t5) {
-			db_fc_threshold = 100000;
-			pr_info("DB Flow Control Disabled.\n");
-		}
-
-		if (!allow_db_coalescing_on_t5) {
-			db_coalescing_threshold = -1;
-			pr_info("DB Coalescing Disabled.\n");
-		}
-	}
-
 	devp = (struct c4iw_dev *)ib_alloc_device(sizeof(*devp));
 	if (!devp) {
 		printk(KERN_ERR MOD "Cannot allocate ib device\n");
@@ -749,6 +749,7 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 	spin_lock_init(&devp->lock);
 	mutex_init(&devp->rdev.stats.lock);
 	mutex_init(&devp->db_mutex);
+	INIT_LIST_HEAD(&devp->db_fc_list);
 
 	if (c4iw_debugfs_root) {
 		devp->debugfs_root = debugfs_create_dir(
@@ -977,13 +978,16 @@ static int disable_qp_db(int id, void *p, void *data)
 
 static void stop_queues(struct uld_ctx *ctx)
 {
-	spin_lock_irq(&ctx->dev->lock);
-	if (ctx->dev->db_state == NORMAL) {
-		ctx->dev->rdev.stats.db_state_transitions++;
-		ctx->dev->db_state = FLOW_CONTROL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&ctx->dev->lock, flags);
+	ctx->dev->rdev.stats.db_state_transitions++;
+	ctx->dev->db_state = STOPPED;
+	if (ctx->dev->rdev.flags & T4_STATUS_PAGE_DISABLED)
 		idr_for_each(&ctx->dev->qpidr, disable_qp_db, NULL);
-	}
-	spin_unlock_irq(&ctx->dev->lock);
+	else
+		ctx->dev->rdev.status_page->db_off = 1;
+	spin_unlock_irqrestore(&ctx->dev->lock, flags);
 }
 
 static int enable_qp_db(int id, void *p, void *data)
@@ -994,15 +998,70 @@ static int enable_qp_db(int id, void *p, void *data)
 	return 0;
 }
 
+static void resume_rc_qp(struct c4iw_qp *qp)
+{
+	spin_lock(&qp->lock);
+	t4_ring_sq_db(&qp->wq, qp->wq.sq.wq_pidx_inc);
+	qp->wq.sq.wq_pidx_inc = 0;
+	t4_ring_rq_db(&qp->wq, qp->wq.rq.wq_pidx_inc);
+	qp->wq.rq.wq_pidx_inc = 0;
+	spin_unlock(&qp->lock);
+}
+
+static void resume_a_chunk(struct uld_ctx *ctx)
+{
+	int i;
+	struct c4iw_qp *qp;
+
+	for (i = 0; i < DB_FC_RESUME_SIZE; i++) {
+		qp = list_first_entry(&ctx->dev->db_fc_list, struct c4iw_qp,
+				      db_fc_entry);
+		list_del_init(&qp->db_fc_entry);
+		resume_rc_qp(qp);
+		if (list_empty(&ctx->dev->db_fc_list))
+			break;
+	}
+}
+
 static void resume_queues(struct uld_ctx *ctx)
 {
 	spin_lock_irq(&ctx->dev->lock);
-	if (ctx->dev->qpcnt <= db_fc_threshold &&
-	    ctx->dev->db_state == FLOW_CONTROL) {
-		ctx->dev->db_state = NORMAL;
-		ctx->dev->rdev.stats.db_state_transitions++;
-		idr_for_each(&ctx->dev->qpidr, enable_qp_db, NULL);
+	if (ctx->dev->db_state != STOPPED)
+		goto out;
+	ctx->dev->db_state = FLOW_CONTROL;
+	while (1) {
+		if (list_empty(&ctx->dev->db_fc_list)) {
+			WARN_ON(ctx->dev->db_state != FLOW_CONTROL);
+			ctx->dev->db_state = NORMAL;
+			ctx->dev->rdev.stats.db_state_transitions++;
+			if (ctx->dev->rdev.flags & T4_STATUS_PAGE_DISABLED) {
+				idr_for_each(&ctx->dev->qpidr, enable_qp_db,
+					     NULL);
+			} else {
+				ctx->dev->rdev.status_page->db_off = 0;
+			}
+			break;
+		} else {
+			if (cxgb4_dbfifo_count(ctx->dev->rdev.lldi.ports[0], 1)
+			    < (ctx->dev->rdev.lldi.dbfifo_int_thresh <<
+			       DB_FC_DRAIN_THRESH)) {
+				resume_a_chunk(ctx);
+			}
+			if (!list_empty(&ctx->dev->db_fc_list)) {
+				spin_unlock_irq(&ctx->dev->lock);
+				if (DB_FC_RESUME_DELAY) {
+					set_current_state(TASK_UNINTERRUPTIBLE);
+					schedule_timeout(DB_FC_RESUME_DELAY);
+				}
+				spin_lock_irq(&ctx->dev->lock);
+				if (ctx->dev->db_state != FLOW_CONTROL)
+					break;
+			}
+		}
 	}
+out:
+	if (ctx->dev->db_state != NORMAL)
+		ctx->dev->rdev.stats.db_fc_interruptions++;
 	spin_unlock_irq(&ctx->dev->lock);
 }
 
@@ -1028,12 +1087,12 @@ static int count_qps(int id, void *p, void *data)
 	return 0;
 }
 
-static void deref_qps(struct qp_list qp_list)
+static void deref_qps(struct qp_list *qp_list)
 {
 	int idx;
 
-	for (idx = 0; idx < qp_list.idx; idx++)
-		c4iw_qp_rem_ref(&qp_list.qps[idx]->ibqp);
+	for (idx = 0; idx < qp_list->idx; idx++)
+		c4iw_qp_rem_ref(&qp_list->qps[idx]->ibqp);
 }
 
 static void recover_lost_dbs(struct uld_ctx *ctx, struct qp_list *qp_list)
@@ -1044,17 +1103,22 @@ static void recover_lost_dbs(struct uld_ctx *ctx, struct qp_list *qp_list)
 	for (idx = 0; idx < qp_list->idx; idx++) {
 		struct c4iw_qp *qp = qp_list->qps[idx];
 
+		spin_lock_irq(&qp->rhp->lock);
+		spin_lock(&qp->lock);
 		ret = cxgb4_sync_txq_pidx(qp->rhp->rdev.lldi.ports[0],
 					  qp->wq.sq.qid,
 					  t4_sq_host_wq_pidx(&qp->wq),
 					  t4_sq_wq_size(&qp->wq));
 		if (ret) {
-			printk(KERN_ERR MOD "%s: Fatal error - "
+			pr_err(KERN_ERR MOD "%s: Fatal error - "
 			       "DB overflow recovery failed - "
 			       "error syncing SQ qid %u\n",
 			       pci_name(ctx->lldi.pdev), qp->wq.sq.qid);
+			spin_unlock(&qp->lock);
+			spin_unlock_irq(&qp->rhp->lock);
 			return;
 		}
+		qp->wq.sq.wq_pidx_inc = 0;
 
 		ret = cxgb4_sync_txq_pidx(qp->rhp->rdev.lldi.ports[0],
 					  qp->wq.rq.qid,
@@ -1062,12 +1126,17 @@ static void recover_lost_dbs(struct uld_ctx *ctx, struct qp_list *qp_list)
 					  t4_rq_wq_size(&qp->wq));
 
 		if (ret) {
-			printk(KERN_ERR MOD "%s: Fatal error - "
+			pr_err(KERN_ERR MOD "%s: Fatal error - "
 			       "DB overflow recovery failed - "
 			       "error syncing RQ qid %u\n",
 			       pci_name(ctx->lldi.pdev), qp->wq.rq.qid);
+			spin_unlock(&qp->lock);
+			spin_unlock_irq(&qp->rhp->lock);
 			return;
 		}
+		qp->wq.rq.wq_pidx_inc = 0;
+		spin_unlock(&qp->lock);
+		spin_unlock_irq(&qp->rhp->lock);
 
 		/* Wait for the dbfifo to drain */
 		while (cxgb4_dbfifo_count(qp->rhp->rdev.lldi.ports[0], 1) > 0) {
@@ -1083,36 +1152,22 @@ static void recover_queues(struct uld_ctx *ctx)
 	struct qp_list qp_list;
 	int ret;
 
-	/* lock out kernel db ringers */
-	mutex_lock(&ctx->dev->db_mutex);
-
-	/* put all queues in to recovery mode */
-	spin_lock_irq(&ctx->dev->lock);
-	ctx->dev->db_state = RECOVERY;
-	ctx->dev->rdev.stats.db_state_transitions++;
-	idr_for_each(&ctx->dev->qpidr, disable_qp_db, NULL);
-	spin_unlock_irq(&ctx->dev->lock);
-
 	/* slow everybody down */
 	set_current_state(TASK_UNINTERRUPTIBLE);
 	schedule_timeout(usecs_to_jiffies(1000));
 
-	/* Wait for the dbfifo to completely drain. */
-	while (cxgb4_dbfifo_count(ctx->dev->rdev.lldi.ports[0], 1) > 0) {
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		schedule_timeout(usecs_to_jiffies(10));
-	}
-
 	/* flush the SGE contexts */
 	ret = cxgb4_flush_eq_cache(ctx->dev->rdev.lldi.ports[0]);
 	if (ret) {
 		printk(KERN_ERR MOD "%s: Fatal error - DB overflow recovery failed\n",
 		       pci_name(ctx->lldi.pdev));
-		goto out;
+		return;
 	}
 
 	/* Count active queues so we can build a list of queues to recover */
 	spin_lock_irq(&ctx->dev->lock);
+	WARN_ON(ctx->dev->db_state != STOPPED);
+	ctx->dev->db_state = RECOVERY;
 	idr_for_each(&ctx->dev->qpidr, count_qps, &count);
 
 	qp_list.qps = kzalloc(count * sizeof *qp_list.qps, GFP_ATOMIC);
@@ -1120,7 +1175,7 @@ static void recover_queues(struct uld_ctx *ctx)
 		printk(KERN_ERR MOD "%s: Fatal error - DB overflow recovery failed\n",
 		       pci_name(ctx->lldi.pdev));
 		spin_unlock_irq(&ctx->dev->lock);
-		goto out;
+		return;
 	}
 	qp_list.idx = 0;
 
@@ -1133,29 +1188,13 @@ static void recover_queues(struct uld_ctx *ctx)
 	recover_lost_dbs(ctx, &qp_list);
 
 	/* we're almost done!  deref the qps and clean up */
-	deref_qps(qp_list);
+	deref_qps(&qp_list);
 	kfree(qp_list.qps);
 
-	/* Wait for the dbfifo to completely drain again */
-	while (cxgb4_dbfifo_count(ctx->dev->rdev.lldi.ports[0], 1) > 0) {
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		schedule_timeout(usecs_to_jiffies(10));
-	}
-
-	/* resume the queues */
 	spin_lock_irq(&ctx->dev->lock);
-	if (ctx->dev->qpcnt > db_fc_threshold)
-		ctx->dev->db_state = FLOW_CONTROL;
-	else {
-		ctx->dev->db_state = NORMAL;
-		idr_for_each(&ctx->dev->qpidr, enable_qp_db, NULL);
-	}
-	ctx->dev->rdev.stats.db_state_transitions++;
+	WARN_ON(ctx->dev->db_state != RECOVERY);
+	ctx->dev->db_state = STOPPED;
 	spin_unlock_irq(&ctx->dev->lock);
-
-out:
-	/* start up kernel db ringers again */
-	mutex_unlock(&ctx->dev->db_mutex);
 }
 
 static int c4iw_uld_control(void *handle, enum cxgb4_control control, ...)
@@ -1165,9 +1204,7 @@ static int c4iw_uld_control(void *handle, enum cxgb4_control control, ...)
 	switch (control) {
 	case CXGB4_CONTROL_DB_FULL:
 		stop_queues(ctx);
-		mutex_lock(&ctx->dev->rdev.stats.lock);
 		ctx->dev->rdev.stats.db_full++;
-		mutex_unlock(&ctx->dev->rdev.stats.lock);
 		break;
 	case CXGB4_CONTROL_DB_EMPTY:
 		resume_queues(ctx);
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index 23eaeabab93b..eb18f9be35e4 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -109,6 +109,7 @@ struct c4iw_dev_ucontext {
 
 enum c4iw_rdev_flags {
 	T4_FATAL_ERROR = (1<<0),
+	T4_STATUS_PAGE_DISABLED = (1<<1),
 };
 
 struct c4iw_stat {
@@ -130,6 +131,7 @@ struct c4iw_stats {
 	u64  db_empty;
 	u64  db_drop;
 	u64  db_state_transitions;
+	u64  db_fc_interruptions;
 	u64  tcam_full;
 	u64  act_ofld_conn_fails;
 	u64  pas_ofld_conn_fails;
@@ -150,6 +152,7 @@ struct c4iw_rdev {
 	unsigned long oc_mw_pa;
 	void __iomem *oc_mw_kva;
 	struct c4iw_stats stats;
+	struct t4_dev_status_page *status_page;
 };
 
 static inline int c4iw_fatal_error(struct c4iw_rdev *rdev)
@@ -211,7 +214,8 @@ static inline int c4iw_wait_for_reply(struct c4iw_rdev *rdev,
 enum db_state {
 	NORMAL = 0,
 	FLOW_CONTROL = 1,
-	RECOVERY = 2
+	RECOVERY = 2,
+	STOPPED = 3
 };
 
 struct c4iw_dev {
@@ -225,10 +229,10 @@ struct c4iw_dev {
 	struct mutex db_mutex;
 	struct dentry *debugfs_root;
 	enum db_state db_state;
-	int qpcnt;
 	struct idr hwtid_idr;
 	struct idr atid_idr;
 	struct idr stid_idr;
+	struct list_head db_fc_list;
 };
 
 static inline struct c4iw_dev *to_c4iw_dev(struct ib_device *ibdev)
@@ -432,6 +436,7 @@ struct c4iw_qp_attributes {
 
 struct c4iw_qp {
 	struct ib_qp ibqp;
+	struct list_head db_fc_entry;
 	struct c4iw_dev *rhp;
 	struct c4iw_ep *ep;
 	struct c4iw_qp_attributes attr;
diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c
index 7e94c9a656a1..e36d2a27c431 100644
--- a/drivers/infiniband/hw/cxgb4/provider.c
+++ b/drivers/infiniband/hw/cxgb4/provider.c
@@ -106,15 +106,54 @@ static struct ib_ucontext *c4iw_alloc_ucontext(struct ib_device *ibdev,
 {
 	struct c4iw_ucontext *context;
 	struct c4iw_dev *rhp = to_c4iw_dev(ibdev);
+	static int warned;
+	struct c4iw_alloc_ucontext_resp uresp;
+	int ret = 0;
+	struct c4iw_mm_entry *mm = NULL;
 
 	PDBG("%s ibdev %p\n", __func__, ibdev);
 	context = kzalloc(sizeof(*context), GFP_KERNEL);
-	if (!context)
-		return ERR_PTR(-ENOMEM);
+	if (!context) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
 	c4iw_init_dev_ucontext(&rhp->rdev, &context->uctx);
 	INIT_LIST_HEAD(&context->mmaps);
 	spin_lock_init(&context->mmap_lock);
+
+	if (udata->outlen < sizeof(uresp)) {
+		if (!warned++)
+			pr_err(MOD "Warning - downlevel libcxgb4 (non-fatal), device status page disabled.");
+		rhp->rdev.flags |= T4_STATUS_PAGE_DISABLED;
+	} else {
+		mm = kmalloc(sizeof(*mm), GFP_KERNEL);
+		if (!mm)
+			goto err_free;
+
+		uresp.status_page_size = PAGE_SIZE;
+
+		spin_lock(&context->mmap_lock);
+		uresp.status_page_key = context->key;
+		context->key += PAGE_SIZE;
+		spin_unlock(&context->mmap_lock);
+
+		ret = ib_copy_to_udata(udata, &uresp, sizeof(uresp));
+		if (ret)
+			goto err_mm;
+
+		mm->key = uresp.status_page_key;
+		mm->addr = virt_to_phys(rhp->rdev.status_page);
+		mm->len = PAGE_SIZE;
+		insert_mmap(context, mm);
+	}
 	return &context->ibucontext;
+err_mm:
+	kfree(mm);
+err_free:
+	kfree(context);
+err:
+	return ERR_PTR(ret);
 }
 
 static int c4iw_mmap(struct ib_ucontext *context, struct vm_area_struct *vma)
diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index 582936708e6e..3b62eb556a47 100644
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -638,6 +638,46 @@ void c4iw_qp_rem_ref(struct ib_qp *qp)
 		wake_up(&(to_c4iw_qp(qp)->wait));
 }
 
+static void add_to_fc_list(struct list_head *head, struct list_head *entry)
+{
+	if (list_empty(entry))
+		list_add_tail(entry, head);
+}
+
+static int ring_kernel_sq_db(struct c4iw_qp *qhp, u16 inc)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&qhp->rhp->lock, flags);
+	spin_lock(&qhp->lock);
+	if (qhp->rhp->db_state == NORMAL) {
+		t4_ring_sq_db(&qhp->wq, inc);
+	} else {
+		add_to_fc_list(&qhp->rhp->db_fc_list, &qhp->db_fc_entry);
+		qhp->wq.sq.wq_pidx_inc += inc;
+	}
+	spin_unlock(&qhp->lock);
+	spin_unlock_irqrestore(&qhp->rhp->lock, flags);
+	return 0;
+}
+
+static int ring_kernel_rq_db(struct c4iw_qp *qhp, u16 inc)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&qhp->rhp->lock, flags);
+	spin_lock(&qhp->lock);
+	if (qhp->rhp->db_state == NORMAL) {
+		t4_ring_rq_db(&qhp->wq, inc);
+	} else {
+		add_to_fc_list(&qhp->rhp->db_fc_list, &qhp->db_fc_entry);
+		qhp->wq.rq.wq_pidx_inc += inc;
+	}
+	spin_unlock(&qhp->lock);
+	spin_unlock_irqrestore(&qhp->rhp->lock, flags);
+	return 0;
+}
+
 int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 		   struct ib_send_wr **bad_wr)
 {
@@ -750,9 +790,13 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 		t4_sq_produce(&qhp->wq, len16);
 		idx += DIV_ROUND_UP(len16*16, T4_EQ_ENTRY_SIZE);
 	}
-	if (t4_wq_db_enabled(&qhp->wq))
+	if (!qhp->rhp->rdev.status_page->db_off) {
 		t4_ring_sq_db(&qhp->wq, idx);
-	spin_unlock_irqrestore(&qhp->lock, flag);
+		spin_unlock_irqrestore(&qhp->lock, flag);
+	} else {
+		spin_unlock_irqrestore(&qhp->lock, flag);
+		ring_kernel_sq_db(qhp, idx);
+	}
 	return err;
 }
 
@@ -812,9 +856,13 @@ int c4iw_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
 		wr = wr->next;
 		num_wrs--;
 	}
-	if (t4_wq_db_enabled(&qhp->wq))
+	if (!qhp->rhp->rdev.status_page->db_off) {
 		t4_ring_rq_db(&qhp->wq, idx);
-	spin_unlock_irqrestore(&qhp->lock, flag);
+		spin_unlock_irqrestore(&qhp->lock, flag);
+	} else {
+		spin_unlock_irqrestore(&qhp->lock, flag);
+		ring_kernel_rq_db(qhp, idx);
+	}
 	return err;
 }
 
@@ -1200,35 +1248,6 @@ out:
 	return ret;
 }
 
-/*
- * Called by the library when the qp has user dbs disabled due to
- * a DB_FULL condition.  This function will single-thread all user
- * DB rings to avoid overflowing the hw db-fifo.
- */
-static int ring_kernel_db(struct c4iw_qp *qhp, u32 qid, u16 inc)
-{
-	int delay = db_delay_usecs;
-
-	mutex_lock(&qhp->rhp->db_mutex);
-	do {
-
-		/*
-		 * The interrupt threshold is dbfifo_int_thresh << 6. So
-		 * make sure we don't cross that and generate an interrupt.
-		 */
-		if (cxgb4_dbfifo_count(qhp->rhp->rdev.lldi.ports[0], 1) <
-		    (qhp->rhp->rdev.lldi.dbfifo_int_thresh << 5)) {
-			writel(QID(qid) | PIDX(inc), qhp->wq.db);
-			break;
-		}
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		schedule_timeout(usecs_to_jiffies(delay));
-		delay = min(delay << 1, 2000);
-	} while (1);
-	mutex_unlock(&qhp->rhp->db_mutex);
-	return 0;
-}
-
 int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp,
 		   enum c4iw_qp_attr_mask mask,
 		   struct c4iw_qp_attributes *attrs,
@@ -1278,11 +1297,11 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp,
 	}
 
 	if (mask & C4IW_QP_ATTR_SQ_DB) {
-		ret = ring_kernel_db(qhp, qhp->wq.sq.qid, attrs->sq_db_inc);
+		ret = ring_kernel_sq_db(qhp, attrs->sq_db_inc);
 		goto out;
 	}
 	if (mask & C4IW_QP_ATTR_RQ_DB) {
-		ret = ring_kernel_db(qhp, qhp->wq.rq.qid, attrs->rq_db_inc);
+		ret = ring_kernel_rq_db(qhp, attrs->rq_db_inc);
 		goto out;
 	}
 
@@ -1465,14 +1484,6 @@ out:
 	return ret;
 }
 
-static int enable_qp_db(int id, void *p, void *data)
-{
-	struct c4iw_qp *qp = p;
-
-	t4_enable_wq_db(&qp->wq);
-	return 0;
-}
-
 int c4iw_destroy_qp(struct ib_qp *ib_qp)
 {
 	struct c4iw_dev *rhp;
@@ -1490,22 +1501,15 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp)
 		c4iw_modify_qp(rhp, qhp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 0);
 	wait_event(qhp->wait, !qhp->ep);
 
-	spin_lock_irq(&rhp->lock);
-	remove_handle_nolock(rhp, &rhp->qpidr, qhp->wq.sq.qid);
-	rhp->qpcnt--;
-	BUG_ON(rhp->qpcnt < 0);
-	if (rhp->qpcnt <= db_fc_threshold && rhp->db_state == FLOW_CONTROL) {
-		rhp->rdev.stats.db_state_transitions++;
-		rhp->db_state = NORMAL;
-		idr_for_each(&rhp->qpidr, enable_qp_db, NULL);
-	}
-	if (db_coalescing_threshold >= 0)
-		if (rhp->qpcnt <= db_coalescing_threshold)
-			cxgb4_enable_db_coalescing(rhp->rdev.lldi.ports[0]);
-	spin_unlock_irq(&rhp->lock);
+	remove_handle(rhp, &rhp->qpidr, qhp->wq.sq.qid);
 	atomic_dec(&qhp->refcnt);
 	wait_event(qhp->wait, !atomic_read(&qhp->refcnt));
 
+	spin_lock_irq(&rhp->lock);
+	if (!list_empty(&qhp->db_fc_entry))
+		list_del_init(&qhp->db_fc_entry);
+	spin_unlock_irq(&rhp->lock);
+
 	ucontext = ib_qp->uobject ?
 		   to_c4iw_ucontext(ib_qp->uobject->context) : NULL;
 	destroy_qp(&rhp->rdev, &qhp->wq,
@@ -1516,14 +1520,6 @@ int c4iw_destroy_qp(struct ib_qp *ib_qp)
 	return 0;
 }
 
-static int disable_qp_db(int id, void *p, void *data)
-{
-	struct c4iw_qp *qp = p;
-
-	t4_disable_wq_db(&qp->wq);
-	return 0;
-}
-
 struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 			     struct ib_udata *udata)
 {
@@ -1610,20 +1606,7 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 	init_waitqueue_head(&qhp->wait);
 	atomic_set(&qhp->refcnt, 1);
 
-	spin_lock_irq(&rhp->lock);
-	if (rhp->db_state != NORMAL)
-		t4_disable_wq_db(&qhp->wq);
-	rhp->qpcnt++;
-	if (rhp->qpcnt > db_fc_threshold && rhp->db_state == NORMAL) {
-		rhp->rdev.stats.db_state_transitions++;
-		rhp->db_state = FLOW_CONTROL;
-		idr_for_each(&rhp->qpidr, disable_qp_db, NULL);
-	}
-	if (db_coalescing_threshold >= 0)
-		if (rhp->qpcnt > db_coalescing_threshold)
-			cxgb4_disable_db_coalescing(rhp->rdev.lldi.ports[0]);
-	ret = insert_handle_nolock(rhp, &rhp->qpidr, qhp, qhp->wq.sq.qid);
-	spin_unlock_irq(&rhp->lock);
+	ret = insert_handle(rhp, &rhp->qpidr, qhp, qhp->wq.sq.qid);
 	if (ret)
 		goto err2;
 
@@ -1709,6 +1692,7 @@ struct ib_qp *c4iw_create_qp(struct ib_pd *pd, struct ib_qp_init_attr *attrs,
 	}
 	qhp->ibqp.qp_num = qhp->wq.sq.qid;
 	init_timer(&(qhp->timer));
+	INIT_LIST_HEAD(&qhp->db_fc_entry);
 	PDBG("%s qhp %p sq_num_entries %d, rq_num_entries %d qpid 0x%0x\n",
 	     __func__, qhp, qhp->attr.sq_num_entries, qhp->attr.rq_num_entries,
 	     qhp->wq.sq.qid);
diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h
index e73ace739183..eeca8b1e6376 100644
--- a/drivers/infiniband/hw/cxgb4/t4.h
+++ b/drivers/infiniband/hw/cxgb4/t4.h
@@ -300,6 +300,7 @@ struct t4_sq {
 	u16 cidx;
 	u16 pidx;
 	u16 wq_pidx;
+	u16 wq_pidx_inc;
 	u16 flags;
 	short flush_cidx;
 };
@@ -324,6 +325,7 @@ struct t4_rq {
 	u16 cidx;
 	u16 pidx;
 	u16 wq_pidx;
+	u16 wq_pidx_inc;
 };
 
 struct t4_wq {
@@ -609,3 +611,7 @@ static inline void t4_set_cq_in_error(struct t4_cq *cq)
 	((struct t4_status_page *)&cq->queue[cq->size])->qp_err = 1;
 }
 #endif
+
+struct t4_dev_status_page {
+	u8 db_off;
+};
diff --git a/drivers/infiniband/hw/cxgb4/user.h b/drivers/infiniband/hw/cxgb4/user.h
index 32b754c35ab7..11ccd276e5d9 100644
--- a/drivers/infiniband/hw/cxgb4/user.h
+++ b/drivers/infiniband/hw/cxgb4/user.h
@@ -70,4 +70,9 @@ struct c4iw_create_qp_resp {
 	__u32 qid_mask;
 	__u32 flags;
 };
+
+struct c4iw_alloc_ucontext_resp {
+	__u64 status_page_key;
+	__u32 status_page_size;
+};
 #endif
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index 50abe1d61287..32db37709263 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -500,6 +500,7 @@ struct sge_txq {
 	spinlock_t db_lock;
 	int db_disabled;
 	unsigned short db_pidx;
+	unsigned short db_pidx_inc;
 	u64 udb;
 };
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index 0ac53dd84c61..cc04d090354c 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -3578,14 +3578,25 @@ static void drain_db_fifo(struct adapter *adap, int usecs)
 
 static void disable_txq_db(struct sge_txq *q)
 {
-	spin_lock_irq(&q->db_lock);
+	unsigned long flags;
+
+	spin_lock_irqsave(&q->db_lock, flags);
 	q->db_disabled = 1;
-	spin_unlock_irq(&q->db_lock);
+	spin_unlock_irqrestore(&q->db_lock, flags);
 }
 
-static void enable_txq_db(struct sge_txq *q)
+static void enable_txq_db(struct adapter *adap, struct sge_txq *q)
 {
 	spin_lock_irq(&q->db_lock);
+	if (q->db_pidx_inc) {
+		/* Make sure that all writes to the TX descriptors
+		 * are committed before we tell HW about them.
+		 */
+		wmb();
+		t4_write_reg(adap, MYPF_REG(SGE_PF_KDOORBELL),
+			     QID(q->cntxt_id) | PIDX(q->db_pidx_inc));
+		q->db_pidx_inc = 0;
+	}
 	q->db_disabled = 0;
 	spin_unlock_irq(&q->db_lock);
 }
@@ -3607,11 +3618,32 @@ static void enable_dbs(struct adapter *adap)
 	int i;
 
 	for_each_ethrxq(&adap->sge, i)
-		enable_txq_db(&adap->sge.ethtxq[i].q);
+		enable_txq_db(adap, &adap->sge.ethtxq[i].q);
 	for_each_ofldrxq(&adap->sge, i)
-		enable_txq_db(&adap->sge.ofldtxq[i].q);
+		enable_txq_db(adap, &adap->sge.ofldtxq[i].q);
 	for_each_port(adap, i)
-		enable_txq_db(&adap->sge.ctrlq[i].q);
+		enable_txq_db(adap, &adap->sge.ctrlq[i].q);
+}
+
+static void notify_rdma_uld(struct adapter *adap, enum cxgb4_control cmd)
+{
+	if (adap->uld_handle[CXGB4_ULD_RDMA])
+		ulds[CXGB4_ULD_RDMA].control(adap->uld_handle[CXGB4_ULD_RDMA],
+				cmd);
+}
+
+static void process_db_full(struct work_struct *work)
+{
+	struct adapter *adap;
+
+	adap = container_of(work, struct adapter, db_full_task);
+
+	drain_db_fifo(adap, dbfifo_drain_delay);
+	enable_dbs(adap);
+	notify_rdma_uld(adap, CXGB4_CONTROL_DB_EMPTY);
+	t4_set_reg_field(adap, SGE_INT_ENABLE3,
+			 DBFIFO_HP_INT | DBFIFO_LP_INT,
+			 DBFIFO_HP_INT | DBFIFO_LP_INT);
 }
 
 static void sync_txq_pidx(struct adapter *adap, struct sge_txq *q)
@@ -3619,7 +3651,7 @@ static void sync_txq_pidx(struct adapter *adap, struct sge_txq *q)
 	u16 hw_pidx, hw_cidx;
 	int ret;
 
-	spin_lock_bh(&q->db_lock);
+	spin_lock_irq(&q->db_lock);
 	ret = read_eq_indices(adap, (u16)q->cntxt_id, &hw_pidx, &hw_cidx);
 	if (ret)
 		goto out;
@@ -3636,7 +3668,8 @@ static void sync_txq_pidx(struct adapter *adap, struct sge_txq *q)
 	}
 out:
 	q->db_disabled = 0;
-	spin_unlock_bh(&q->db_lock);
+	q->db_pidx_inc = 0;
+	spin_unlock_irq(&q->db_lock);
 	if (ret)
 		CH_WARN(adap, "DB drop recovery failed.\n");
 }
@@ -3652,29 +3685,6 @@ static void recover_all_queues(struct adapter *adap)
 		sync_txq_pidx(adap, &adap->sge.ctrlq[i].q);
 }
 
-static void notify_rdma_uld(struct adapter *adap, enum cxgb4_control cmd)
-{
-	mutex_lock(&uld_mutex);
-	if (adap->uld_handle[CXGB4_ULD_RDMA])
-		ulds[CXGB4_ULD_RDMA].control(adap->uld_handle[CXGB4_ULD_RDMA],
-				cmd);
-	mutex_unlock(&uld_mutex);
-}
-
-static void process_db_full(struct work_struct *work)
-{
-	struct adapter *adap;
-
-	adap = container_of(work, struct adapter, db_full_task);
-
-	notify_rdma_uld(adap, CXGB4_CONTROL_DB_FULL);
-	drain_db_fifo(adap, dbfifo_drain_delay);
-	t4_set_reg_field(adap, SGE_INT_ENABLE3,
-			 DBFIFO_HP_INT | DBFIFO_LP_INT,
-			 DBFIFO_HP_INT | DBFIFO_LP_INT);
-	notify_rdma_uld(adap, CXGB4_CONTROL_DB_EMPTY);
-}
-
 static void process_db_drop(struct work_struct *work)
 {
 	struct adapter *adap;
@@ -3682,11 +3692,13 @@ static void process_db_drop(struct work_struct *work)
 	adap = container_of(work, struct adapter, db_drop_task);
 
 	if (is_t4(adap->params.chip)) {
-		disable_dbs(adap);
+		drain_db_fifo(adap, dbfifo_drain_delay);
 		notify_rdma_uld(adap, CXGB4_CONTROL_DB_DROP);
-		drain_db_fifo(adap, 1);
+		drain_db_fifo(adap, dbfifo_drain_delay);
 		recover_all_queues(adap);
+		drain_db_fifo(adap, dbfifo_drain_delay);
 		enable_dbs(adap);
+		notify_rdma_uld(adap, CXGB4_CONTROL_DB_EMPTY);
 	} else {
 		u32 dropped_db = t4_read_reg(adap, 0x010ac);
 		u16 qid = (dropped_db >> 15) & 0x1ffff;
@@ -3727,6 +3739,8 @@ static void process_db_drop(struct work_struct *work)
 void t4_db_full(struct adapter *adap)
 {
 	if (is_t4(adap->params.chip)) {
+		disable_dbs(adap);
+		notify_rdma_uld(adap, CXGB4_CONTROL_DB_FULL);
 		t4_set_reg_field(adap, SGE_INT_ENABLE3,
 				 DBFIFO_HP_INT | DBFIFO_LP_INT, 0);
 		queue_work(workq, &adap->db_full_task);
@@ -3735,8 +3749,11 @@ void t4_db_full(struct adapter *adap)
 
 void t4_db_dropped(struct adapter *adap)
 {
-	if (is_t4(adap->params.chip))
-		queue_work(workq, &adap->db_drop_task);
+	if (is_t4(adap->params.chip)) {
+		disable_dbs(adap);
+		notify_rdma_uld(adap, CXGB4_CONTROL_DB_FULL);
+	}
+	queue_work(workq, &adap->db_drop_task);
 }
 
 static void uld_attach(struct adapter *adap, unsigned int uld)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index 46429f9d0592..d4db382ff8c7 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -860,9 +860,10 @@ static void cxgb_pio_copy(u64 __iomem *dst, u64 *src)
 static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
 {
 	unsigned int *wr, index;
+	unsigned long flags;
 
 	wmb();            /* write descriptors before telling HW */
-	spin_lock(&q->db_lock);
+	spin_lock_irqsave(&q->db_lock, flags);
 	if (!q->db_disabled) {
 		if (is_t4(adap->params.chip)) {
 			t4_write_reg(adap, MYPF_REG(SGE_PF_KDOORBELL),
@@ -878,9 +879,10 @@ static inline void ring_tx_db(struct adapter *adap, struct sge_txq *q, int n)
 				writel(n,  adap->bar2 + q->udb + 8);
 			wmb();
 		}
-	}
+	} else
+		q->db_pidx_inc += n;
 	q->db_pidx = q->pidx;
-	spin_unlock(&q->db_lock);
+	spin_unlock_irqrestore(&q->db_lock, flags);
 }
 
 /**