--- zzzz-none-000/linux-3.10.107/drivers/infiniband/hw/cxgb4/qp.c	2017-06-27 09:49:32.000000000 +0000
+++ scorpion-7490-727/linux-3.10.107/drivers/infiniband/hw/cxgb4/qp.c	2021-02-04 17:41:59.000000000 +0000
@@ -58,6 +58,31 @@
 module_param(max_fr_immd, int, 0644);
 MODULE_PARM_DESC(max_fr_immd, "fastreg threshold for using DSGL instead of immedate");
 
+static int alloc_ird(struct c4iw_dev *dev, u32 ird)
+{
+	int ret = 0;
+
+	spin_lock_irq(&dev->lock);
+	if (ird <= dev->avail_ird)
+		dev->avail_ird -= ird;
+	else
+		ret = -ENOMEM;
+	spin_unlock_irq(&dev->lock);
+
+	if (ret)
+		dev_warn(&dev->rdev.lldi.pdev->dev,
+			 "device IRD resources exhausted\n");
+
+	return ret;
+}
+
+static void free_ird(struct c4iw_dev *dev, int ird)
+{
+	spin_lock_irq(&dev->lock);
+	dev->avail_ird += ird;
+	spin_unlock_irq(&dev->lock);
+}
+
 static void set_state(struct c4iw_qp *qhp, enum c4iw_qp_state state)
 {
 	unsigned long flag;
@@ -140,6 +165,33 @@
 	return 0;
 }
 
+/*
+ * Determine the BAR2 virtual address and qid. If pbar2_pa is not NULL,
+ * then this is a user mapping so compute the page-aligned physical address
+ * for mapping.
+ */
+void __iomem *c4iw_bar2_addrs(struct c4iw_rdev *rdev, unsigned int qid,
+			      enum cxgb4_bar2_qtype qtype,
+			      unsigned int *pbar2_qid, u64 *pbar2_pa)
+{
+	u64 bar2_qoffset;
+	int ret;
+
+	ret = cxgb4_bar2_sge_qregs(rdev->lldi.ports[0], qid, qtype,
+				   pbar2_pa ? 1 : 0,
+				   &bar2_qoffset, pbar2_qid);
+	if (ret)
+		return NULL;
+
+	if (pbar2_pa)
+		*pbar2_pa = (rdev->bar2_pa + bar2_qoffset) & PAGE_MASK;
+
+	if (is_t4(rdev->lldi.adapter_type))
+		return NULL;
+
+	return rdev->bar2_kva + bar2_qoffset;
+}
+
 static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq,
 		     struct t4_cq *rcq, struct t4_cq *scq,
 		     struct c4iw_dev_ucontext *uctx)
@@ -180,9 +232,9 @@
 	}
 
 	/*
-	 * RQT must be a power of 2.
+	 * RQT must be a power of 2 and at least 16 deep.
 	 */
-	wq->rq.rqt_size = roundup_pow_of_two(wq->rq.size);
+	wq->rq.rqt_size = roundup_pow_of_two(max_t(u16, wq->rq.size, 16));
 	wq->rq.rqt_hwaddr = c4iw_rqtpool_alloc(rdev, wq->rq.rqt_size);
 	if (!wq->rq.rqt_hwaddr) {
 		ret = -ENOMEM;
@@ -211,15 +263,23 @@
 	dma_unmap_addr_set(&wq->rq, mapping, wq->rq.dma_addr);
 
 	wq->db = rdev->lldi.db_reg;
-	wq->gts = rdev->lldi.gts_reg;
-	if (user) {
-		wq->sq.udb = (u64)pci_resource_start(rdev->lldi.pdev, 2) +
-					(wq->sq.qid << rdev->qpshift);
-		wq->sq.udb &= PAGE_MASK;
-		wq->rq.udb = (u64)pci_resource_start(rdev->lldi.pdev, 2) +
-					(wq->rq.qid << rdev->qpshift);
-		wq->rq.udb &= PAGE_MASK;
+
+	wq->sq.bar2_va = c4iw_bar2_addrs(rdev, wq->sq.qid, T4_BAR2_QTYPE_EGRESS,
+					 &wq->sq.bar2_qid,
+					 user ? &wq->sq.bar2_pa : NULL);
+	wq->rq.bar2_va = c4iw_bar2_addrs(rdev, wq->rq.qid, T4_BAR2_QTYPE_EGRESS,
+					 &wq->rq.bar2_qid,
+					 user ? &wq->rq.bar2_pa : NULL);
+
+	/*
+	 * User mode must have bar2 access.
+	 */
+	if (user && (!wq->sq.bar2_pa || !wq->rq.bar2_pa)) {
+		pr_warn(MOD "%s: sqid %u or rqid %u not in BAR2 range.\n",
+			pci_name(rdev->lldi.pdev), wq->sq.qid, wq->rq.qid);
+		goto free_dma;
 	}
+
 	wq->rdev = rdev;
 	wq->rq.msn = 1;
 
@@ -236,11 +296,11 @@
 	res_wr = (struct fw_ri_res_wr *)__skb_put(skb, wr_len);
 	memset(res_wr, 0, wr_len);
 	res_wr->op_nres = cpu_to_be32(
-			FW_WR_OP(FW_RI_RES_WR) |
-			V_FW_RI_RES_WR_NRES(2) |
-			FW_WR_COMPL(1));
+			FW_WR_OP_V(FW_RI_RES_WR) |
+			FW_RI_RES_WR_NRES_V(2) |
+			FW_WR_COMPL_F);
 	res_wr->len16_pkd = cpu_to_be32(DIV_ROUND_UP(wr_len, 16));
-	res_wr->cookie = (unsigned long) &wr_wait;
+	res_wr->cookie = (uintptr_t)&wr_wait;
 	res = res_wr->res;
 	res->u.sqrq.restype = FW_RI_RES_TYPE_SQ;
 	res->u.sqrq.op = FW_RI_RES_OP_WRITE;
@@ -248,22 +308,23 @@
 	/*
 	 * eqsize is the number of 64B entries plus the status page size.
 	 */
-	eqsize = wq->sq.size * T4_SQ_NUM_SLOTS + T4_EQ_STATUS_ENTRIES;
+	eqsize = wq->sq.size * T4_SQ_NUM_SLOTS +
+		rdev->hw_queue.t4_eq_status_entries;
 
 	res->u.sqrq.fetchszm_to_iqid = cpu_to_be32(
-		V_FW_RI_RES_WR_HOSTFCMODE(0) |	/* no host cidx updates */
-		V_FW_RI_RES_WR_CPRIO(0) |	/* don't keep in chip cache */
-		V_FW_RI_RES_WR_PCIECHN(0) |	/* set by uP at ri_init time */
-		(t4_sq_onchip(&wq->sq) ? F_FW_RI_RES_WR_ONCHIP : 0) |
-		V_FW_RI_RES_WR_IQID(scq->cqid));
+		FW_RI_RES_WR_HOSTFCMODE_V(0) |	/* no host cidx updates */
+		FW_RI_RES_WR_CPRIO_V(0) |	/* don't keep in chip cache */
+		FW_RI_RES_WR_PCIECHN_V(0) |	/* set by uP at ri_init time */
+		(t4_sq_onchip(&wq->sq) ? FW_RI_RES_WR_ONCHIP_F : 0) |
+		FW_RI_RES_WR_IQID_V(scq->cqid));
 	res->u.sqrq.dcaen_to_eqsize = cpu_to_be32(
-		V_FW_RI_RES_WR_DCAEN(0) |
-		V_FW_RI_RES_WR_DCACPU(0) |
-		V_FW_RI_RES_WR_FBMIN(2) |
-		V_FW_RI_RES_WR_FBMAX(2) |
-		V_FW_RI_RES_WR_CIDXFTHRESHO(0) |
-		V_FW_RI_RES_WR_CIDXFTHRESH(0) |
-		V_FW_RI_RES_WR_EQSIZE(eqsize));
+		FW_RI_RES_WR_DCAEN_V(0) |
+		FW_RI_RES_WR_DCACPU_V(0) |
+		FW_RI_RES_WR_FBMIN_V(2) |
+		FW_RI_RES_WR_FBMAX_V(2) |
+		FW_RI_RES_WR_CIDXFTHRESHO_V(0) |
+		FW_RI_RES_WR_CIDXFTHRESH_V(0) |
+		FW_RI_RES_WR_EQSIZE_V(eqsize));
 	res->u.sqrq.eqid = cpu_to_be32(wq->sq.qid);
 	res->u.sqrq.eqaddr = cpu_to_be64(wq->sq.dma_addr);
 	res++;
@@ -273,20 +334,21 @@
 	/*
 	 * eqsize is the number of 64B entries plus the status page size.
 	 */
-	eqsize = wq->rq.size * T4_RQ_NUM_SLOTS + T4_EQ_STATUS_ENTRIES;
+	eqsize = wq->rq.size * T4_RQ_NUM_SLOTS +
+		rdev->hw_queue.t4_eq_status_entries;
 	res->u.sqrq.fetchszm_to_iqid = cpu_to_be32(
-		V_FW_RI_RES_WR_HOSTFCMODE(0) |	/* no host cidx updates */
-		V_FW_RI_RES_WR_CPRIO(0) |	/* don't keep in chip cache */
-		V_FW_RI_RES_WR_PCIECHN(0) |	/* set by uP at ri_init time */
-		V_FW_RI_RES_WR_IQID(rcq->cqid));
+		FW_RI_RES_WR_HOSTFCMODE_V(0) |	/* no host cidx updates */
+		FW_RI_RES_WR_CPRIO_V(0) |	/* don't keep in chip cache */
+		FW_RI_RES_WR_PCIECHN_V(0) |	/* set by uP at ri_init time */
+		FW_RI_RES_WR_IQID_V(rcq->cqid));
 	res->u.sqrq.dcaen_to_eqsize = cpu_to_be32(
-		V_FW_RI_RES_WR_DCAEN(0) |
-		V_FW_RI_RES_WR_DCACPU(0) |
-		V_FW_RI_RES_WR_FBMIN(2) |
-		V_FW_RI_RES_WR_FBMAX(2) |
-		V_FW_RI_RES_WR_CIDXFTHRESHO(0) |
-		V_FW_RI_RES_WR_CIDXFTHRESH(0) |
-		V_FW_RI_RES_WR_EQSIZE(eqsize));
+		FW_RI_RES_WR_DCAEN_V(0) |
+		FW_RI_RES_WR_DCACPU_V(0) |
+		FW_RI_RES_WR_FBMIN_V(2) |
+		FW_RI_RES_WR_FBMAX_V(2) |
+		FW_RI_RES_WR_CIDXFTHRESHO_V(0) |
+		FW_RI_RES_WR_CIDXFTHRESH_V(0) |
+		FW_RI_RES_WR_EQSIZE_V(eqsize));
 	res->u.sqrq.eqid = cpu_to_be32(wq->rq.qid);
 	res->u.sqrq.eqaddr = cpu_to_be64(wq->rq.dma_addr);
 
@@ -299,9 +361,9 @@
 	if (ret)
 		goto free_dma;
 
-	PDBG("%s sqid 0x%x rqid 0x%x kdb 0x%p squdb 0x%llx rqudb 0x%llx\n",
+	PDBG("%s sqid 0x%x rqid 0x%x kdb 0x%p sq_bar2_addr %p rq_bar2_addr %p\n",
 	     __func__, wq->sq.qid, wq->rq.qid, wq->db,
-	     (unsigned long long)wq->sq.udb, (unsigned long long)wq->rq.udb);
+	     wq->sq.bar2_va, wq->rq.bar2_va);
 
 	return 0;
 free_dma:
@@ -406,25 +468,27 @@
 	case IB_WR_SEND:
 		if (wr->send_flags & IB_SEND_SOLICITED)
 			wqe->send.sendop_pkd = cpu_to_be32(
-				V_FW_RI_SEND_WR_SENDOP(FW_RI_SEND_WITH_SE));
+				FW_RI_SEND_WR_SENDOP_V(FW_RI_SEND_WITH_SE));
 		else
 			wqe->send.sendop_pkd = cpu_to_be32(
-				V_FW_RI_SEND_WR_SENDOP(FW_RI_SEND));
+				FW_RI_SEND_WR_SENDOP_V(FW_RI_SEND));
 		wqe->send.stag_inv = 0;
 		break;
 	case IB_WR_SEND_WITH_INV:
 		if (wr->send_flags & IB_SEND_SOLICITED)
 			wqe->send.sendop_pkd = cpu_to_be32(
-				V_FW_RI_SEND_WR_SENDOP(FW_RI_SEND_WITH_SE_INV));
+				FW_RI_SEND_WR_SENDOP_V(FW_RI_SEND_WITH_SE_INV));
 		else
 			wqe->send.sendop_pkd = cpu_to_be32(
-				V_FW_RI_SEND_WR_SENDOP(FW_RI_SEND_WITH_INV));
+				FW_RI_SEND_WR_SENDOP_V(FW_RI_SEND_WITH_INV));
 		wqe->send.stag_inv = cpu_to_be32(wr->ex.invalidate_rkey);
 		break;
 
 	default:
 		return -EINVAL;
 	}
+	wqe->send.r3 = 0;
+	wqe->send.r4 = 0;
 
 	plen = 0;
 	if (wr->num_sge) {
@@ -468,8 +532,8 @@
 	if (wr->num_sge > T4_MAX_SEND_SGE)
 		return -EINVAL;
 	wqe->write.r2 = 0;
-	wqe->write.stag_sink = cpu_to_be32(wr->wr.rdma.rkey);
-	wqe->write.to_sink = cpu_to_be64(wr->wr.rdma.remote_addr);
+	wqe->write.stag_sink = cpu_to_be32(rdma_wr(wr)->rkey);
+	wqe->write.to_sink = cpu_to_be64(rdma_wr(wr)->remote_addr);
 	if (wr->num_sge) {
 		if (wr->send_flags & IB_SEND_INLINE) {
 			ret = build_immd(sq, wqe->write.u.immd_src, wr,
@@ -506,10 +570,10 @@
 	if (wr->num_sge > 1)
 		return -EINVAL;
 	if (wr->num_sge) {
-		wqe->read.stag_src = cpu_to_be32(wr->wr.rdma.rkey);
-		wqe->read.to_src_hi = cpu_to_be32((u32)(wr->wr.rdma.remote_addr
+		wqe->read.stag_src = cpu_to_be32(rdma_wr(wr)->rkey);
+		wqe->read.to_src_hi = cpu_to_be32((u32)(rdma_wr(wr)->remote_addr
 							>> 32));
-		wqe->read.to_src_lo = cpu_to_be32((u32)wr->wr.rdma.remote_addr);
+		wqe->read.to_src_lo = cpu_to_be32((u32)rdma_wr(wr)->remote_addr);
 		wqe->read.stag_sink = cpu_to_be32(wr->sg_list[0].lkey);
 		wqe->read.plen = cpu_to_be32(wr->sg_list[0].length);
 		wqe->read.to_sink_hi = cpu_to_be32((u32)(wr->sg_list[0].addr
@@ -545,46 +609,41 @@
 	return 0;
 }
 
-static int build_fastreg(struct t4_sq *sq, union t4_wr *wqe,
-			 struct ib_send_wr *wr, u8 *len16, u8 t5dev)
+static int build_memreg(struct t4_sq *sq, union t4_wr *wqe,
+			struct ib_reg_wr *wr, u8 *len16, u8 t5dev)
 {
-
+	struct c4iw_mr *mhp = to_c4iw_mr(wr->mr);
 	struct fw_ri_immd *imdp;
 	__be64 *p;
 	int i;
-	int pbllen = roundup(wr->wr.fast_reg.page_list_len * sizeof(u64), 32);
+	int pbllen = roundup(mhp->mpl_len * sizeof(u64), 32);
 	int rem;
 
-	if (wr->wr.fast_reg.page_list_len > T4_MAX_FR_DEPTH)
+	if (mhp->mpl_len > t4_max_fr_depth(use_dsgl))
 		return -EINVAL;
 
 	wqe->fr.qpbinde_to_dcacpu = 0;
-	wqe->fr.pgsz_shift = wr->wr.fast_reg.page_shift - 12;
+	wqe->fr.pgsz_shift = ilog2(wr->mr->page_size) - 12;
 	wqe->fr.addr_type = FW_RI_VA_BASED_TO;
-	wqe->fr.mem_perms = c4iw_ib_to_tpt_access(wr->wr.fast_reg.access_flags);
+	wqe->fr.mem_perms = c4iw_ib_to_tpt_access(wr->access);
 	wqe->fr.len_hi = 0;
-	wqe->fr.len_lo = cpu_to_be32(wr->wr.fast_reg.length);
-	wqe->fr.stag = cpu_to_be32(wr->wr.fast_reg.rkey);
-	wqe->fr.va_hi = cpu_to_be32(wr->wr.fast_reg.iova_start >> 32);
-	wqe->fr.va_lo_fbo = cpu_to_be32(wr->wr.fast_reg.iova_start &
+	wqe->fr.len_lo = cpu_to_be32(mhp->ibmr.length);
+	wqe->fr.stag = cpu_to_be32(wr->key);
+	wqe->fr.va_hi = cpu_to_be32(mhp->ibmr.iova >> 32);
+	wqe->fr.va_lo_fbo = cpu_to_be32(mhp->ibmr.iova &
 					0xffffffff);
 
 	if (t5dev && use_dsgl && (pbllen > max_fr_immd)) {
-		struct c4iw_fr_page_list *c4pl =
-			to_c4iw_fr_page_list(wr->wr.fast_reg.page_list);
 		struct fw_ri_dsgl *sglp;
 
-		for (i = 0; i < wr->wr.fast_reg.page_list_len; i++) {
-			wr->wr.fast_reg.page_list->page_list[i] = (__force u64)
-				cpu_to_be64((u64)
-				wr->wr.fast_reg.page_list->page_list[i]);
-		}
+		for (i = 0; i < mhp->mpl_len; i++)
+			mhp->mpl[i] = (__force u64)cpu_to_be64((u64)mhp->mpl[i]);
 
 		sglp = (struct fw_ri_dsgl *)(&wqe->fr + 1);
 		sglp->op = FW_RI_DATA_DSGL;
 		sglp->r1 = 0;
 		sglp->nsge = cpu_to_be16(1);
-		sglp->addr0 = cpu_to_be64(c4pl->dma_addr);
+		sglp->addr0 = cpu_to_be64(mhp->mpl_addr);
 		sglp->len0 = cpu_to_be32(pbllen);
 
 		*len16 = DIV_ROUND_UP(sizeof(wqe->fr) + sizeof(*sglp), 16);
@@ -596,9 +655,8 @@
 		imdp->immdlen = cpu_to_be32(pbllen);
 		p = (__be64 *)(imdp + 1);
 		rem = pbllen;
-		for (i = 0; i < wr->wr.fast_reg.page_list_len; i++) {
-			*p = cpu_to_be64(
-				(u64)wr->wr.fast_reg.page_list->page_list[i]);
+		for (i = 0; i < mhp->mpl_len; i++) {
+			*p = cpu_to_be64((u64)mhp->mpl[i]);
 			rem -= sizeof(*p);
 			if (++p == (__be64 *)&sq->queue[sq->size])
 				p = (__be64 *)sq->queue;
@@ -638,6 +696,46 @@
 		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, NULL);
+	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, NULL);
+	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)
 {
@@ -646,7 +744,7 @@
 	enum fw_wr_opcodes fw_opcode = 0;
 	enum fw_ri_wr_flags fw_flags;
 	struct c4iw_qp *qhp;
-	union t4_wr *wqe;
+	union t4_wr *wqe = NULL;
 	u32 num_wrs;
 	struct t4_swsqe *swsqe;
 	unsigned long flag;
@@ -675,7 +773,7 @@
 		fw_flags = 0;
 		if (wr->send_flags & IB_SEND_SOLICITED)
 			fw_flags |= FW_RI_SOLICITED_EVENT_FLAG;
-		if (wr->send_flags & IB_SEND_SIGNALED)
+		if (wr->send_flags & IB_SEND_SIGNALED || qhp->sq_sig_all)
 			fw_flags |= FW_RI_COMPLETION_FLAG;
 		swsqe = &qhp->wq.sq.sw_sq[qhp->wq.sq.pidx];
 		switch (wr->opcode) {
@@ -710,13 +808,13 @@
 			if (!qhp->wq.sq.oldest_read)
 				qhp->wq.sq.oldest_read = swsqe;
 			break;
-		case IB_WR_FAST_REG_MR:
+		case IB_WR_REG_MR:
 			fw_opcode = FW_RI_FR_NSMR_WR;
 			swsqe->opcode = FW_RI_FAST_REGISTER;
-			err = build_fastreg(&qhp->wq.sq, wqe, wr, &len16,
-					    is_t5(
-					    qhp->rhp->rdev.lldi.adapter_type) ?
-					    1 : 0);
+			err = build_memreg(&qhp->wq.sq, wqe, reg_wr(wr), &len16,
+					   is_t5(
+					   qhp->rhp->rdev.lldi.adapter_type) ?
+					   1 : 0);
 			break;
 		case IB_WR_LOCAL_INV:
 			if (wr->send_flags & IB_SEND_FENCE)
@@ -736,8 +834,15 @@
 		}
 		swsqe->idx = qhp->wq.sq.pidx;
 		swsqe->complete = 0;
-		swsqe->signaled = (wr->send_flags & IB_SEND_SIGNALED);
+		swsqe->signaled = (wr->send_flags & IB_SEND_SIGNALED) ||
+				  qhp->sq_sig_all;
+		swsqe->flushed = 0;
 		swsqe->wr_id = wr->wr_id;
+		if (c4iw_wr_log) {
+			swsqe->sge_ts = cxgb4_read_sge_timestamp(
+					qhp->rhp->rdev.lldi.ports[0]);
+			getnstimeofday(&swsqe->host_ts);
+		}
 
 		init_wr_hdr(wqe, qhp->wq.sq.pidx, fw_opcode, fw_flags, len16);
 
@@ -749,9 +854,13 @@
 		t4_sq_produce(&qhp->wq, len16);
 		idx += DIV_ROUND_UP(len16*16, T4_EQ_ENTRY_SIZE);
 	}
-	if (t4_wq_db_enabled(&qhp->wq))
-		t4_ring_sq_db(&qhp->wq, idx);
-	spin_unlock_irqrestore(&qhp->lock, flag);
+	if (!qhp->rhp->rdev.status_page->db_off) {
+		t4_ring_sq_db(&qhp->wq, idx, wqe);
+		spin_unlock_irqrestore(&qhp->lock, flag);
+	} else {
+		spin_unlock_irqrestore(&qhp->lock, flag);
+		ring_kernel_sq_db(qhp, idx);
+	}
 	return err;
 }
 
@@ -760,7 +869,7 @@
 {
 	int err = 0;
 	struct c4iw_qp *qhp;
-	union t4_recv_wr *wqe;
+	union t4_recv_wr *wqe = NULL;
 	u32 num_wrs;
 	u8 len16 = 0;
 	unsigned long flag;
@@ -796,6 +905,13 @@
 		}
 
 		qhp->wq.rq.sw_rq[qhp->wq.rq.pidx].wr_id = wr->wr_id;
+		if (c4iw_wr_log) {
+			qhp->wq.rq.sw_rq[qhp->wq.rq.pidx].sge_ts =
+				cxgb4_read_sge_timestamp(
+						qhp->rhp->rdev.lldi.ports[0]);
+			getnstimeofday(
+				&qhp->wq.rq.sw_rq[qhp->wq.rq.pidx].host_ts);
+		}
 
 		wqe->recv.opcode = FW_RI_RECV_WR;
 		wqe->recv.r1 = 0;
@@ -811,9 +927,13 @@
 		wr = wr->next;
 		num_wrs--;
 	}
-	if (t4_wq_db_enabled(&qhp->wq))
-		t4_ring_rq_db(&qhp->wq, idx);
-	spin_unlock_irqrestore(&qhp->lock, flag);
+	if (!qhp->rhp->rdev.status_page->db_off) {
+		t4_ring_rq_db(&qhp->wq, idx, wqe);
+		spin_unlock_irqrestore(&qhp->lock, flag);
+	} else {
+		spin_unlock_irqrestore(&qhp->lock, flag);
+		ring_kernel_rq_db(qhp, idx);
+	}
 	return err;
 }
 
@@ -975,10 +1095,10 @@
 
 	wqe = (struct fw_ri_wr *)__skb_put(skb, sizeof(*wqe));
 	memset(wqe, 0, sizeof *wqe);
-	wqe->op_compl = cpu_to_be32(FW_WR_OP(FW_RI_INIT_WR));
+	wqe->op_compl = cpu_to_be32(FW_WR_OP_V(FW_RI_INIT_WR));
 	wqe->flowid_len16 = cpu_to_be32(
-		FW_WR_FLOWID(qhp->ep->hwtid) |
-		FW_WR_LEN16(DIV_ROUND_UP(sizeof *wqe, 16)));
+		FW_WR_FLOWID_V(qhp->ep->hwtid) |
+		FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*wqe), 16)));
 
 	wqe->u.terminate.type = FW_RI_TYPE_TERMINATE;
 	wqe->u.terminate.immdlen = cpu_to_be32(sizeof *term);
@@ -998,7 +1118,7 @@
 		       struct c4iw_cq *schp)
 {
 	int count;
-	int flushed;
+	int rq_flushed, sq_flushed;
 	unsigned long flag;
 
 	PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp);
@@ -1006,29 +1126,50 @@
 	/* locking hierarchy: cq lock first, then qp lock. */
 	spin_lock_irqsave(&rchp->lock, flag);
 	spin_lock(&qhp->lock);
-	c4iw_flush_hw_cq(&rchp->cq);
+
+	if (qhp->wq.flushed) {
+		spin_unlock(&qhp->lock);
+		spin_unlock_irqrestore(&rchp->lock, flag);
+		return;
+	}
+	qhp->wq.flushed = 1;
+
+	c4iw_flush_hw_cq(rchp);
 	c4iw_count_rcqes(&rchp->cq, &qhp->wq, &count);
-	flushed = c4iw_flush_rq(&qhp->wq, &rchp->cq, count);
+	rq_flushed = c4iw_flush_rq(&qhp->wq, &rchp->cq, count);
 	spin_unlock(&qhp->lock);
 	spin_unlock_irqrestore(&rchp->lock, flag);
-	if (flushed) {
-		spin_lock_irqsave(&rchp->comp_handler_lock, flag);
-		(*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context);
-		spin_unlock_irqrestore(&rchp->comp_handler_lock, flag);
-	}
 
 	/* locking hierarchy: cq lock first, then qp lock. */
 	spin_lock_irqsave(&schp->lock, flag);
 	spin_lock(&qhp->lock);
-	c4iw_flush_hw_cq(&schp->cq);
-	c4iw_count_scqes(&schp->cq, &qhp->wq, &count);
-	flushed = c4iw_flush_sq(&qhp->wq, &schp->cq, count);
+	if (schp != rchp)
+		c4iw_flush_hw_cq(schp);
+	sq_flushed = c4iw_flush_sq(qhp);
 	spin_unlock(&qhp->lock);
 	spin_unlock_irqrestore(&schp->lock, flag);
-	if (flushed) {
-		spin_lock_irqsave(&schp->comp_handler_lock, flag);
-		(*schp->ibcq.comp_handler)(&schp->ibcq, schp->ibcq.cq_context);
-		spin_unlock_irqrestore(&schp->comp_handler_lock, flag);
+
+	if (schp == rchp) {
+		if (t4_clear_cq_armed(&rchp->cq) &&
+		    (rq_flushed || sq_flushed)) {
+			spin_lock_irqsave(&rchp->comp_handler_lock, flag);
+			(*rchp->ibcq.comp_handler)(&rchp->ibcq,
+						   rchp->ibcq.cq_context);
+			spin_unlock_irqrestore(&rchp->comp_handler_lock, flag);
+		}
+	} else {
+		if (t4_clear_cq_armed(&rchp->cq) && rq_flushed) {
+			spin_lock_irqsave(&rchp->comp_handler_lock, flag);
+			(*rchp->ibcq.comp_handler)(&rchp->ibcq,
+						   rchp->ibcq.cq_context);
+			spin_unlock_irqrestore(&rchp->comp_handler_lock, flag);
+		}
+		if (t4_clear_cq_armed(&schp->cq) && sq_flushed) {
+			spin_lock_irqsave(&schp->comp_handler_lock, flag);
+			(*schp->ibcq.comp_handler)(&schp->ibcq,
+						   schp->ibcq.cq_context);
+			spin_unlock_irqrestore(&schp->comp_handler_lock, flag);
+		}
 	}
 }
 
@@ -1037,11 +1178,11 @@
 	struct c4iw_cq *rchp, *schp;
 	unsigned long flag;
 
-	rchp = get_chp(qhp->rhp, qhp->attr.rcq);
-	schp = get_chp(qhp->rhp, qhp->attr.scq);
+	rchp = to_c4iw_cq(qhp->ibqp.recv_cq);
+	schp = to_c4iw_cq(qhp->ibqp.send_cq);
 
+	t4_set_wq_in_error(&qhp->wq);
 	if (qhp->ibqp.uobject) {
-		t4_set_wq_in_error(&qhp->wq);
 		t4_set_cq_in_error(&rchp->cq);
 		spin_lock_irqsave(&rchp->comp_handler_lock, flag);
 		(*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context);
@@ -1076,12 +1217,12 @@
 	wqe = (struct fw_ri_wr *)__skb_put(skb, sizeof(*wqe));
 	memset(wqe, 0, sizeof *wqe);
 	wqe->op_compl = cpu_to_be32(
-		FW_WR_OP(FW_RI_INIT_WR) |
-		FW_WR_COMPL(1));
+		FW_WR_OP_V(FW_RI_INIT_WR) |
+		FW_WR_COMPL_F);
 	wqe->flowid_len16 = cpu_to_be32(
-		FW_WR_FLOWID(ep->hwtid) |
-		FW_WR_LEN16(DIV_ROUND_UP(sizeof *wqe, 16)));
-	wqe->cookie = (unsigned long) &ep->com.wr_wait;
+		FW_WR_FLOWID_V(ep->hwtid) |
+		FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*wqe), 16)));
+	wqe->cookie = (uintptr_t)&ep->com.wr_wait;
 
 	wqe->u.fini.type = FW_RI_TYPE_FINI;
 	ret = c4iw_ofld_send(&rhp->rdev, skb);
@@ -1126,29 +1267,37 @@
 	int ret;
 	struct sk_buff *skb;
 
-	PDBG("%s qhp %p qid 0x%x tid %u\n", __func__, qhp, qhp->wq.sq.qid,
-	     qhp->ep->hwtid);
+	PDBG("%s qhp %p qid 0x%x tid %u ird %u ord %u\n", __func__, qhp,
+	     qhp->wq.sq.qid, qhp->ep->hwtid, qhp->ep->ird, qhp->ep->ord);
 
 	skb = alloc_skb(sizeof *wqe, GFP_KERNEL);
-	if (!skb)
-		return -ENOMEM;
+	if (!skb) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	ret = alloc_ird(rhp, qhp->attr.max_ird);
+	if (ret) {
+		qhp->attr.max_ird = 0;
+		kfree_skb(skb);
+		goto out;
+	}
 	set_wr_txq(skb, CPL_PRIORITY_DATA, qhp->ep->txq_idx);
 
 	wqe = (struct fw_ri_wr *)__skb_put(skb, sizeof(*wqe));
 	memset(wqe, 0, sizeof *wqe);
 	wqe->op_compl = cpu_to_be32(
-		FW_WR_OP(FW_RI_INIT_WR) |
-		FW_WR_COMPL(1));
+		FW_WR_OP_V(FW_RI_INIT_WR) |
+		FW_WR_COMPL_F);
 	wqe->flowid_len16 = cpu_to_be32(
-		FW_WR_FLOWID(qhp->ep->hwtid) |
-		FW_WR_LEN16(DIV_ROUND_UP(sizeof *wqe, 16)));
+		FW_WR_FLOWID_V(qhp->ep->hwtid) |
+		FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*wqe), 16)));
 
-	wqe->cookie = (unsigned long) &qhp->ep->com.wr_wait;
+	wqe->cookie = (uintptr_t)&qhp->ep->com.wr_wait;
 
 	wqe->u.init.type = FW_RI_TYPE_INIT;
 	wqe->u.init.mpareqbit_p2ptype =
-		V_FW_RI_WR_MPAREQBIT(qhp->attr.mpa_attr.initiator) |
-		V_FW_RI_WR_P2PTYPE(qhp->attr.mpa_attr.p2p_type);
+		FW_RI_WR_MPAREQBIT_V(qhp->attr.mpa_attr.initiator) |
+		FW_RI_WR_P2PTYPE_V(qhp->attr.mpa_attr.p2p_type);
 	wqe->u.init.mpa_attrs = FW_RI_MPA_IETF_ENABLE;
 	if (qhp->attr.mpa_attr.recv_marker_enabled)
 		wqe->u.init.mpa_attrs |= FW_RI_MPA_RX_MARKER_ENABLE;
@@ -1182,44 +1331,19 @@
 
 	ret = c4iw_ofld_send(&rhp->rdev, skb);
 	if (ret)
-		goto out;
+		goto err1;
 
 	ret = c4iw_wait_for_reply(&rhp->rdev, &qhp->ep->com.wr_wait,
 				  qhp->ep->hwtid, qhp->wq.sq.qid, __func__);
+	if (!ret)
+		goto out;
+err1:
+	free_ird(rhp, qhp->attr.max_ird);
 out:
 	PDBG("%s ret %d\n", __func__, ret);
 	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,
@@ -1259,7 +1383,7 @@
 			newattr.max_ord = attrs->max_ord;
 		}
 		if (mask & C4IW_QP_ATTR_MAX_IRD) {
-			if (attrs->max_ird > c4iw_max_read_depth) {
+			if (attrs->max_ird > cur_max_read_depth(rhp)) {
 				ret = -EINVAL;
 				goto out;
 			}
@@ -1269,11 +1393,11 @@
 	}
 
 	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;
 	}
 
@@ -1323,6 +1447,7 @@
 		switch (attrs->next_state) {
 		case C4IW_QP_STATE_CLOSING:
 			BUG_ON(atomic_read(&qhp->ep->com.kref.refcount) < 2);
+			t4_set_wq_in_error(&qhp->wq);
 			set_state(qhp, C4IW_QP_STATE_CLOSING);
 			ep = qhp->ep;
 			if (!internal) {
@@ -1330,28 +1455,30 @@
 				disconnect = 1;
 				c4iw_get_ep(&qhp->ep->com);
 			}
-			if (qhp->ibqp.uobject)
-				t4_set_wq_in_error(&qhp->wq);
 			ret = rdma_fini(rhp, qhp, ep);
 			if (ret)
 				goto err;
 			break;
 		case C4IW_QP_STATE_TERMINATE:
+			t4_set_wq_in_error(&qhp->wq);
 			set_state(qhp, C4IW_QP_STATE_TERMINATE);
 			qhp->attr.layer_etype = attrs->layer_etype;
 			qhp->attr.ecode = attrs->ecode;
-			if (qhp->ibqp.uobject)
-				t4_set_wq_in_error(&qhp->wq);
 			ep = qhp->ep;
-			if (!internal)
+			if (!internal) {
+				c4iw_get_ep(&qhp->ep->com);
 				terminate = 1;
-			disconnect = 1;
-			c4iw_get_ep(&qhp->ep->com);
+				disconnect = 1;
+			} else {
+				terminate = qhp->attr.send_term;
+				ret = rdma_fini(rhp, qhp, ep);
+				if (ret)
+					goto err;
+			}
 			break;
 		case C4IW_QP_STATE_ERROR:
+			t4_set_wq_in_error(&qhp->wq);
 			set_state(qhp, C4IW_QP_STATE_ERROR);
-			if (qhp->ibqp.uobject)
-				t4_set_wq_in_error(&qhp->wq);
 			if (!internal) {
 				abort = 1;
 				disconnect = 1;
@@ -1424,9 +1551,9 @@
 	set_state(qhp, C4IW_QP_STATE_ERROR);
 	free = 1;
 	abort = 1;
-	wake_up(&qhp->wait);
 	BUG_ON(!ep);
 	flush_qp(qhp);
+	wake_up(&qhp->wait);
 out:
 	mutex_unlock(&qhp->mutex);
 
@@ -1454,14 +1581,6 @@
 	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;
@@ -1479,22 +1598,16 @@
 		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);
+	free_ird(rhp, qhp->attr.max_ird);
+
 	ucontext = ib_qp->uobject ?
 		   to_c4iw_ucontext(ib_qp->uobject->context) : NULL;
 	destroy_qp(&rhp->rdev, &qhp->wq,
@@ -1505,14 +1618,6 @@
 	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)
 {
@@ -1522,7 +1627,7 @@
 	struct c4iw_cq *schp;
 	struct c4iw_cq *rchp;
 	struct c4iw_create_qp_resp uresp;
-	int sqsize, rqsize;
+	unsigned int sqsize, rqsize;
 	struct c4iw_ucontext *ucontext;
 	int ret;
 	struct c4iw_mm_entry *mm1, *mm2, *mm3, *mm4, *mm5 = NULL;
@@ -1542,33 +1647,38 @@
 	if (attrs->cap.max_inline_data > T4_MAX_SEND_INLINE)
 		return ERR_PTR(-EINVAL);
 
-	rqsize = roundup(attrs->cap.max_recv_wr + 1, 16);
-	if (rqsize > T4_MAX_RQ_SIZE)
+	if (attrs->cap.max_recv_wr > rhp->rdev.hw_queue.t4_max_rq_size)
 		return ERR_PTR(-E2BIG);
+	rqsize = attrs->cap.max_recv_wr + 1;
+	if (rqsize < 8)
+		rqsize = 8;
 
-	sqsize = roundup(attrs->cap.max_send_wr + 1, 16);
-	if (sqsize > T4_MAX_SQ_SIZE)
+	if (attrs->cap.max_send_wr > rhp->rdev.hw_queue.t4_max_sq_size)
 		return ERR_PTR(-E2BIG);
+	sqsize = attrs->cap.max_send_wr + 1;
+	if (sqsize < 8)
+		sqsize = 8;
 
 	ucontext = pd->uobject ? to_c4iw_ucontext(pd->uobject->context) : NULL;
 
-
 	qhp = kzalloc(sizeof(*qhp), GFP_KERNEL);
 	if (!qhp)
 		return ERR_PTR(-ENOMEM);
 	qhp->wq.sq.size = sqsize;
-	qhp->wq.sq.memsize = (sqsize + 1) * sizeof *qhp->wq.sq.queue;
+	qhp->wq.sq.memsize =
+		(sqsize + rhp->rdev.hw_queue.t4_eq_status_entries) *
+		sizeof(*qhp->wq.sq.queue) + 16 * sizeof(__be64);
+	qhp->wq.sq.flush_cidx = -1;
 	qhp->wq.rq.size = rqsize;
-	qhp->wq.rq.memsize = (rqsize + 1) * sizeof *qhp->wq.rq.queue;
+	qhp->wq.rq.memsize =
+		(rqsize + rhp->rdev.hw_queue.t4_eq_status_entries) *
+		sizeof(*qhp->wq.rq.queue);
 
 	if (ucontext) {
 		qhp->wq.sq.memsize = roundup(qhp->wq.sq.memsize, PAGE_SIZE);
 		qhp->wq.rq.memsize = roundup(qhp->wq.rq.memsize, PAGE_SIZE);
 	}
 
-	PDBG("%s sqsize %u sqmemsize %zu rqsize %u rqmemsize %zu\n",
-	     __func__, sqsize, qhp->wq.sq.memsize, rqsize, qhp->wq.rq.memsize);
-
 	ret = create_qp(&rhp->rdev, &qhp->wq, &schp->cq, &rchp->cq,
 			ucontext ? &ucontext->uctx : &rhp->rdev.uctx);
 	if (ret)
@@ -1592,27 +1702,15 @@
 	qhp->attr.enable_rdma_read = 1;
 	qhp->attr.enable_rdma_write = 1;
 	qhp->attr.enable_bind = 1;
-	qhp->attr.max_ord = 1;
-	qhp->attr.max_ird = 1;
+	qhp->attr.max_ord = 0;
+	qhp->attr.max_ird = 0;
+	qhp->sq_sig_all = attrs->sq_sig_type == IB_SIGNAL_ALL_WR;
 	spin_lock_init(&qhp->lock);
 	mutex_init(&qhp->mutex);
 	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;
 
@@ -1657,6 +1755,8 @@
 		if (mm5) {
 			uresp.ma_sync_key = ucontext->key;
 			ucontext->key += PAGE_SIZE;
+		} else {
+			uresp.ma_sync_key =  0;
 		}
 		uresp.sq_key = ucontext->key;
 		ucontext->key += PAGE_SIZE;
@@ -1679,26 +1779,29 @@
 		mm2->len = PAGE_ALIGN(qhp->wq.rq.memsize);
 		insert_mmap(ucontext, mm2);
 		mm3->key = uresp.sq_db_gts_key;
-		mm3->addr = qhp->wq.sq.udb;
+		mm3->addr = (__force unsigned long)qhp->wq.sq.bar2_pa;
 		mm3->len = PAGE_SIZE;
 		insert_mmap(ucontext, mm3);
 		mm4->key = uresp.rq_db_gts_key;
-		mm4->addr = qhp->wq.rq.udb;
+		mm4->addr = (__force unsigned long)qhp->wq.rq.bar2_pa;
 		mm4->len = PAGE_SIZE;
 		insert_mmap(ucontext, mm4);
 		if (mm5) {
 			mm5->key = uresp.ma_sync_key;
 			mm5->addr = (pci_resource_start(rhp->rdev.lldi.pdev, 0)
-				    + A_PCIE_MA_SYNC) & PAGE_MASK;
+				    + PCIE_MA_SYNC_A) & PAGE_MASK;
 			mm5->len = PAGE_SIZE;
 			insert_mmap(ucontext, mm5);
 		}
 	}
 	qhp->ibqp.qp_num = qhp->wq.sq.qid;
 	init_timer(&(qhp->timer));
-	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);
+	INIT_LIST_HEAD(&qhp->db_fc_entry);
+	PDBG("%s sq id %u size %u memsize %zu num_entries %u "
+	     "rq id %u size %u memsize %zu num_entries %u\n", __func__,
+	     qhp->wq.sq.qid, qhp->wq.sq.size, qhp->wq.sq.memsize,
+	     attrs->cap.max_send_wr, qhp->wq.rq.qid, qhp->wq.rq.size,
+	     qhp->wq.rq.memsize, attrs->cap.max_recv_wr);
 	return &qhp->ibqp;
 err8:
 	kfree(mm5);
@@ -1759,11 +1862,15 @@
 	/*
 	 * Use SQ_PSN and RQ_PSN to pass in IDX_INC values for
 	 * ringing the queue db when we're in DB_FULL mode.
+	 * Only allow this on T4 devices.
 	 */
 	attrs.sq_db_inc = attr->sq_psn;
 	attrs.rq_db_inc = attr->rq_psn;
 	mask |= (attr_mask & IB_QP_SQ_PSN) ? C4IW_QP_ATTR_SQ_DB : 0;
 	mask |= (attr_mask & IB_QP_RQ_PSN) ? C4IW_QP_ATTR_RQ_DB : 0;
+	if (!is_t4(to_c4iw_qp(ibqp)->rhp->rdev.lldi.adapter_type) &&
+	    (mask & (C4IW_QP_ATTR_SQ_DB|C4IW_QP_ATTR_RQ_DB)))
+		return -EINVAL;
 
 	return c4iw_modify_qp(rhp, qhp, mask, &attrs, 0);
 }
@@ -1782,5 +1889,11 @@
 	memset(attr, 0, sizeof *attr);
 	memset(init_attr, 0, sizeof *init_attr);
 	attr->qp_state = to_ib_qp_state(qhp->attr.state);
+	init_attr->cap.max_send_wr = qhp->attr.sq_num_entries;
+	init_attr->cap.max_recv_wr = qhp->attr.rq_num_entries;
+	init_attr->cap.max_send_sge = qhp->attr.sq_max_sges;
+	init_attr->cap.max_recv_sge = qhp->attr.sq_max_sges;
+	init_attr->cap.max_inline_data = T4_MAX_SEND_INLINE;
+	init_attr->sq_sig_type = qhp->sq_sig_all ? IB_SIGNAL_ALL_WR : 0;
 	return 0;
 }