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

Pull rdma updates from Jason Gunthorpe:
 "This has been a slightly more active cycle than normal with ongoing
  core changes and quite a lot of collected driver updates.

   - Various driver fixes for bnxt_re, cxgb4, hns, mlx5, pvrdma, rxe

   - A new data transfer mode for HFI1 giving higher performance

   - Significant functional and bug fix update to the mlx5
     On-Demand-Paging MR feature

   - A chip hang reset recovery system for hns

   - Change mm->pinned_vm to an atomic64

   - Update bnxt_re to support a new 57500 chip

   - A sane netlink 'rdma link add' method for creating rxe devices and
     fixing the various unregistration race conditions in rxe's
     unregister flow

   - Allow lookup up objects by an ID over netlink

   - Various reworking of the core to driver interface:
       - drivers should not assume umem SGLs are in PAGE_SIZE chunks
       - ucontext is accessed via udata not other means
       - start to make the core code responsible for object memory
         allocation
       - drivers should convert struct device to struct ib_device via a
         helper
       - drivers have more tools to avoid use after unregister problems"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (280 commits)
  net/mlx5: ODP support for XRC transport is not enabled by default in FW
  IB/hfi1: Close race condition on user context disable and close
  RDMA/umem: Revert broken 'off by one' fix
  RDMA/umem: minor bug fix in error handling path
  RDMA/hns: Use GFP_ATOMIC in hns_roce_v2_modify_qp
  cxgb4: kfree mhp after the debug print
  IB/rdmavt: Fix concurrency panics in QP post_send and modify to error
  IB/rdmavt: Fix loopback send with invalidate ordering
  IB/iser: Fix dma_nents type definition
  IB/mlx5: Set correct write permissions for implicit ODP MR
  bnxt_re: Clean cq for kernel consumers only
  RDMA/uverbs: Don't do double free of allocated PD
  RDMA: Handle ucontext allocations by IB/core
  RDMA/core: Fix a WARN() message
  bnxt_re: fix the regression due to changes in alloc_pbl
  IB/mlx4: Increase the timeout for CM cache
  IB/core: Abort page fault handler silently during owning process exit
  IB/mlx5: Validate correct PD before prefetch MR
  IB/mlx5: Protect against prefetch of invalid MR
  RDMA/uverbs: Store PR pointer before it is overwritten
  ...

14 files changed, 604 insertions, 358 deletions

diff --git a/drivers/infiniband/hw/bnxt_re/Kconfig b/drivers/infiniband/hw/bnxt_re/Kconfig
index 19982a4a9bba..d25439c305f7 100644
--- a/drivers/infiniband/hw/bnxt_re/Kconfig
+++ b/drivers/infiniband/hw/bnxt_re/Kconfig
@@ -1,5 +1,6 @@
 config INFINIBAND_BNXT_RE
     tristate "Broadcom Netxtreme HCA support"
+    depends on 64BIT
     depends on ETHERNET && NETDEVICES && PCI && INET && DCB
     select NET_VENDOR_BROADCOM
     select BNXT
diff --git a/drivers/infiniband/hw/bnxt_re/Makefile b/drivers/infiniband/hw/bnxt_re/Makefile
index 6e3bc25cc140..ee9bb1be61ea 100644
--- a/drivers/infiniband/hw/bnxt_re/Makefile
+++ b/drivers/infiniband/hw/bnxt_re/Makefile
@@ -1,6 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0
 
-ccflags-y := -Idrivers/net/ethernet/broadcom/bnxt
+ccflags-y := -I $(srctree)/drivers/net/ethernet/broadcom/bnxt
 obj-$(CONFIG_INFINIBAND_BNXT_RE) += bnxt_re.o
 bnxt_re-y := main.o ib_verbs.o \
 	     qplib_res.o qplib_rcfw.o	\
diff --git a/drivers/infiniband/hw/bnxt_re/bnxt_re.h b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
index 31baa8939a4f..e55a1666c0cd 100644
--- a/drivers/infiniband/hw/bnxt_re/bnxt_re.h
+++ b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
@@ -124,6 +124,7 @@ struct bnxt_re_dev {
 #define BNXT_RE_FLAG_ISSUE_ROCE_STATS          29
 	struct net_device		*netdev;
 	unsigned int			version, major, minor;
+	struct bnxt_qplib_chip_ctx	chip_ctx;
 	struct bnxt_en_dev		*en_dev;
 	struct bnxt_msix_entry		msix_entries[BNXT_RE_MAX_MSIX];
 	int				num_msix;
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
index 1e2515e2eb62..071b2fc38b0b 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c
@@ -48,6 +48,7 @@
 #include <rdma/ib_addr.h>
 #include <rdma/ib_mad.h>
 #include <rdma/ib_cache.h>
+#include <rdma/uverbs_ioctl.h>
 
 #include "bnxt_ulp.h"
 
@@ -563,41 +564,29 @@ fail:
 }
 
 /* Protection Domains */
-int bnxt_re_dealloc_pd(struct ib_pd *ib_pd)
+void bnxt_re_dealloc_pd(struct ib_pd *ib_pd)
 {
 	struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd);
 	struct bnxt_re_dev *rdev = pd->rdev;
-	int rc;
 
 	bnxt_re_destroy_fence_mr(pd);
 
-	if (pd->qplib_pd.id) {
-		rc = bnxt_qplib_dealloc_pd(&rdev->qplib_res,
-					   &rdev->qplib_res.pd_tbl,
-					   &pd->qplib_pd);
-		if (rc)
-			dev_err(rdev_to_dev(rdev), "Failed to deallocate HW PD");
-	}
-
-	kfree(pd);
-	return 0;
+	if (pd->qplib_pd.id)
+		bnxt_qplib_dealloc_pd(&rdev->qplib_res, &rdev->qplib_res.pd_tbl,
+				      &pd->qplib_pd);
 }
 
-struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev,
-			       struct ib_ucontext *ucontext,
-			       struct ib_udata *udata)
+int bnxt_re_alloc_pd(struct ib_pd *ibpd, struct ib_ucontext *ucontext,
+		     struct ib_udata *udata)
 {
+	struct ib_device *ibdev = ibpd->device;
 	struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev);
 	struct bnxt_re_ucontext *ucntx = container_of(ucontext,
 						      struct bnxt_re_ucontext,
 						      ib_uctx);
-	struct bnxt_re_pd *pd;
+	struct bnxt_re_pd *pd = container_of(ibpd, struct bnxt_re_pd, ib_pd);
 	int rc;
 
-	pd = kzalloc(sizeof(*pd), GFP_KERNEL);
-	if (!pd)
-		return ERR_PTR(-ENOMEM);
-
 	pd->rdev = rdev;
 	if (bnxt_qplib_alloc_pd(&rdev->qplib_res.pd_tbl, &pd->qplib_pd)) {
 		dev_err(rdev_to_dev(rdev), "Failed to allocate HW PD");
@@ -637,13 +626,12 @@ struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev,
 		if (bnxt_re_create_fence_mr(pd))
 			dev_warn(rdev_to_dev(rdev),
 				 "Failed to create Fence-MR\n");
-	return &pd->ib_pd;
+	return 0;
 dbfail:
-	(void)bnxt_qplib_dealloc_pd(&rdev->qplib_res, &rdev->qplib_res.pd_tbl,
-				    &pd->qplib_pd);
+	bnxt_qplib_dealloc_pd(&rdev->qplib_res, &rdev->qplib_res.pd_tbl,
+			      &pd->qplib_pd);
 fail:
-	kfree(pd);
-	return ERR_PTR(rc);
+	return rc;
 }
 
 /* Address Handles */
@@ -663,17 +651,36 @@ int bnxt_re_destroy_ah(struct ib_ah *ib_ah, u32 flags)
 	return 0;
 }
 
+static u8 bnxt_re_stack_to_dev_nw_type(enum rdma_network_type ntype)
+{
+	u8 nw_type;
+
+	switch (ntype) {
+	case RDMA_NETWORK_IPV4:
+		nw_type = CMDQ_CREATE_AH_TYPE_V2IPV4;
+		break;
+	case RDMA_NETWORK_IPV6:
+		nw_type = CMDQ_CREATE_AH_TYPE_V2IPV6;
+		break;
+	default:
+		nw_type = CMDQ_CREATE_AH_TYPE_V1;
+		break;
+	}
+	return nw_type;
+}
+
 struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd,
 				struct rdma_ah_attr *ah_attr,
 				u32 flags,
 				struct ib_udata *udata)
 {
 	struct bnxt_re_pd *pd = container_of(ib_pd, struct bnxt_re_pd, ib_pd);
+	const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr);
 	struct bnxt_re_dev *rdev = pd->rdev;
+	const struct ib_gid_attr *sgid_attr;
 	struct bnxt_re_ah *ah;
-	const struct ib_global_route *grh = rdma_ah_read_grh(ah_attr);
-	int rc;
 	u8 nw_type;
+	int rc;
 
 	if (!(rdma_ah_get_ah_flags(ah_attr) & IB_AH_GRH)) {
 		dev_err(rdev_to_dev(rdev), "Failed to alloc AH: GRH not set");
@@ -700,28 +707,11 @@ struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd,
 	ah->qplib_ah.flow_label = grh->flow_label;
 	ah->qplib_ah.hop_limit = grh->hop_limit;
 	ah->qplib_ah.sl = rdma_ah_get_sl(ah_attr);
-	if (udata &&
-	    !rdma_is_multicast_addr((struct in6_addr *)
-				    grh->dgid.raw) &&
-	    !rdma_link_local_addr((struct in6_addr *)
-				  grh->dgid.raw)) {
-		const struct ib_gid_attr *sgid_attr;
 
-		sgid_attr = grh->sgid_attr;
-		/* Get network header type for this GID */
-		nw_type = rdma_gid_attr_network_type(sgid_attr);
-		switch (nw_type) {
-		case RDMA_NETWORK_IPV4:
-			ah->qplib_ah.nw_type = CMDQ_CREATE_AH_TYPE_V2IPV4;
-			break;
-		case RDMA_NETWORK_IPV6:
-			ah->qplib_ah.nw_type = CMDQ_CREATE_AH_TYPE_V2IPV6;
-			break;
-		default:
-			ah->qplib_ah.nw_type = CMDQ_CREATE_AH_TYPE_V1;
-			break;
-		}
-	}
+	sgid_attr = grh->sgid_attr;
+	/* Get network header type for this GID */
+	nw_type = rdma_gid_attr_network_type(sgid_attr);
+	ah->qplib_ah.nw_type = bnxt_re_stack_to_dev_nw_type(nw_type);
 
 	memcpy(ah->qplib_ah.dmac, ah_attr->roce.dmac, ETH_ALEN);
 	rc = bnxt_qplib_create_ah(&rdev->qplib_res, &ah->qplib_ah,
@@ -733,12 +723,11 @@ struct ib_ah *bnxt_re_create_ah(struct ib_pd *ib_pd,
 
 	/* Write AVID to shared page. */
 	if (udata) {
-		struct ib_ucontext *ib_uctx = ib_pd->uobject->context;
-		struct bnxt_re_ucontext *uctx;
+		struct bnxt_re_ucontext *uctx = rdma_udata_to_drv_context(
+			udata, struct bnxt_re_ucontext, ib_uctx);
 		unsigned long flag;
 		u32 *wrptr;
 
-		uctx = container_of(ib_uctx, struct bnxt_re_ucontext, ib_uctx);
 		spin_lock_irqsave(&uctx->sh_lock, flag);
 		wrptr = (u32 *)(uctx->shpg + BNXT_RE_AVID_OFFT);
 		*wrptr = ah->qplib_ah.id;
@@ -804,8 +793,8 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp)
 {
 	struct bnxt_re_qp *qp = container_of(ib_qp, struct bnxt_re_qp, ib_qp);
 	struct bnxt_re_dev *rdev = qp->rdev;
-	int rc;
 	unsigned int flags;
+	int rc;
 
 	bnxt_qplib_flush_cqn_wq(&qp->qplib_qp);
 	rc = bnxt_qplib_destroy_qp(&rdev->qplib_res, &qp->qplib_qp);
@@ -814,9 +803,12 @@ int bnxt_re_destroy_qp(struct ib_qp *ib_qp)
 		return rc;
 	}
 
-	flags = bnxt_re_lock_cqs(qp);
-	bnxt_qplib_clean_qp(&qp->qplib_qp);
-	bnxt_re_unlock_cqs(qp, flags);
+	if (rdma_is_kernel_res(&qp->ib_qp.res)) {
+		flags = bnxt_re_lock_cqs(qp);
+		bnxt_qplib_clean_qp(&qp->qplib_qp);
+		bnxt_re_unlock_cqs(qp, flags);
+	}
+
 	bnxt_qplib_free_qp_res(&rdev->qplib_res, &qp->qplib_qp);
 
 	if (ib_qp->qp_type == IB_QPT_GSI && rdev->qp1_sqp) {
@@ -882,21 +874,23 @@ static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd,
 	struct bnxt_re_qp_req ureq;
 	struct bnxt_qplib_qp *qplib_qp = &qp->qplib_qp;
 	struct ib_umem *umem;
-	int bytes = 0;
-	struct ib_ucontext *context = pd->ib_pd.uobject->context;
-	struct bnxt_re_ucontext *cntx = container_of(context,
-						     struct bnxt_re_ucontext,
-						     ib_uctx);
+	int bytes = 0, psn_sz;
+	struct bnxt_re_ucontext *cntx = rdma_udata_to_drv_context(
+		udata, struct bnxt_re_ucontext, ib_uctx);
+
 	if (ib_copy_from_udata(&ureq, udata, sizeof(ureq)))
 		return -EFAULT;
 
 	bytes = (qplib_qp->sq.max_wqe * BNXT_QPLIB_MAX_SQE_ENTRY_SIZE);
 	/* Consider mapping PSN search memory only for RC QPs. */
-	if (qplib_qp->type == CMDQ_CREATE_QP_TYPE_RC)
-		bytes += (qplib_qp->sq.max_wqe * sizeof(struct sq_psn_search));
+	if (qplib_qp->type == CMDQ_CREATE_QP_TYPE_RC) {
+		psn_sz = bnxt_qplib_is_chip_gen_p5(&rdev->chip_ctx) ?
+					sizeof(struct sq_psn_search_ext) :
+					sizeof(struct sq_psn_search);
+		bytes += (qplib_qp->sq.max_wqe * psn_sz);
+	}
 	bytes = PAGE_ALIGN(bytes);
-	umem = ib_umem_get(context, ureq.qpsva, bytes,
-			   IB_ACCESS_LOCAL_WRITE, 1);
+	umem = ib_umem_get(udata, ureq.qpsva, bytes, IB_ACCESS_LOCAL_WRITE, 1);
 	if (IS_ERR(umem))
 		return PTR_ERR(umem);
 
@@ -908,7 +902,7 @@ static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd,
 	if (!qp->qplib_qp.srq) {
 		bytes = (qplib_qp->rq.max_wqe * BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
 		bytes = PAGE_ALIGN(bytes);
-		umem = ib_umem_get(context, ureq.qprva, bytes,
+		umem = ib_umem_get(udata, ureq.qprva, bytes,
 				   IB_ACCESS_LOCAL_WRITE, 1);
 		if (IS_ERR(umem))
 			goto rqfail;
@@ -1066,12 +1060,17 @@ struct ib_qp *bnxt_re_create_qp(struct ib_pd *ib_pd,
 	qp->qplib_qp.pd = &pd->qplib_pd;
 	qp->qplib_qp.qp_handle = (u64)(unsigned long)(&qp->qplib_qp);
 	qp->qplib_qp.type = __from_ib_qp_type(qp_init_attr->qp_type);
+
+	if (qp_init_attr->qp_type == IB_QPT_GSI &&
+	    bnxt_qplib_is_chip_gen_p5(&rdev->chip_ctx))
+		qp->qplib_qp.type = CMDQ_CREATE_QP_TYPE_GSI;
 	if (qp->qplib_qp.type == IB_QPT_MAX) {
 		dev_err(rdev_to_dev(rdev), "QP type 0x%x not supported",
 			qp->qplib_qp.type);
 		rc = -EINVAL;
 		goto fail;
 	}
+
 	qp->qplib_qp.max_inline_data = qp_init_attr->cap.max_inline_data;
 	qp->qplib_qp.sig_type = ((qp_init_attr->sq_sig_type ==
 				  IB_SIGNAL_ALL_WR) ? true : false);
@@ -1132,7 +1131,8 @@ struct ib_qp *bnxt_re_create_qp(struct ib_pd *ib_pd,
 
 	qp->qplib_qp.mtu = ib_mtu_enum_to_int(iboe_get_mtu(rdev->netdev->mtu));
 
-	if (qp_init_attr->qp_type == IB_QPT_GSI) {
+	if (qp_init_attr->qp_type == IB_QPT_GSI &&
+	    !(bnxt_qplib_is_chip_gen_p5(&rdev->chip_ctx))) {
 		/* Allocate 1 more than what's provided */
 		entries = roundup_pow_of_two(qp_init_attr->cap.max_send_wr + 1);
 		qp->qplib_qp.sq.max_wqe = min_t(u32, entries,
@@ -1361,17 +1361,15 @@ static int bnxt_re_init_user_srq(struct bnxt_re_dev *rdev,
 	struct bnxt_qplib_srq *qplib_srq = &srq->qplib_srq;
 	struct ib_umem *umem;
 	int bytes = 0;
-	struct ib_ucontext *context = pd->ib_pd.uobject->context;
-	struct bnxt_re_ucontext *cntx = container_of(context,
-						     struct bnxt_re_ucontext,
-						     ib_uctx);
+	struct bnxt_re_ucontext *cntx = rdma_udata_to_drv_context(
+		udata, struct bnxt_re_ucontext, ib_uctx);
+
 	if (ib_copy_from_udata(&ureq, udata, sizeof(ureq)))
 		return -EFAULT;
 
 	bytes = (qplib_srq->max_wqe * BNXT_QPLIB_MAX_RQE_ENTRY_SIZE);
 	bytes = PAGE_ALIGN(bytes);
-	umem = ib_umem_get(context, ureq.srqva, bytes,
-			   IB_ACCESS_LOCAL_WRITE, 1);
+	umem = ib_umem_get(udata, ureq.srqva, bytes, IB_ACCESS_LOCAL_WRITE, 1);
 	if (IS_ERR(umem))
 		return PTR_ERR(umem);
 
@@ -1646,6 +1644,9 @@ int bnxt_re_modify_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr,
 			__from_ib_access_flags(qp_attr->qp_access_flags);
 		/* LOCAL_WRITE access must be set to allow RC receive */
 		qp->qplib_qp.access |= BNXT_QPLIB_ACCESS_LOCAL_WRITE;
+		/* Temp: Set all params on QP as of now */
+		qp->qplib_qp.access |= CMDQ_MODIFY_QP_ACCESS_REMOTE_WRITE;
+		qp->qplib_qp.access |= CMDQ_MODIFY_QP_ACCESS_REMOTE_READ;
 	}
 	if (qp_attr_mask & IB_QP_PKEY_INDEX) {
 		qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_PKEY;
@@ -2093,7 +2094,8 @@ static int bnxt_re_build_qp1_shadow_qp_recv(struct bnxt_re_qp *qp,
 
 static int is_ud_qp(struct bnxt_re_qp *qp)
 {
-	return qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_UD;
+	return (qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_UD ||
+		qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_GSI);
 }
 
 static int bnxt_re_build_send_wqe(struct bnxt_re_qp *qp,
@@ -2397,7 +2399,7 @@ int bnxt_re_post_send(struct ib_qp *ib_qp, const struct ib_send_wr *wr,
 		switch (wr->opcode) {
 		case IB_WR_SEND:
 		case IB_WR_SEND_WITH_IMM:
-			if (ib_qp->qp_type == IB_QPT_GSI) {
+			if (qp->qplib_qp.type == CMDQ_CREATE_QP1_TYPE_GSI) {
 				rc = bnxt_re_build_qp1_send_v2(qp, wr, &wqe,
 							       payload_sz);
 				if (rc)
@@ -2527,7 +2529,8 @@ int bnxt_re_post_recv(struct ib_qp *ib_qp, const struct ib_recv_wr *wr,
 		wqe.wr_id = wr->wr_id;
 		wqe.type = BNXT_QPLIB_SWQE_TYPE_RECV;
 
-		if (ib_qp->qp_type == IB_QPT_GSI)
+		if (ib_qp->qp_type == IB_QPT_GSI &&
+		    qp->qplib_qp.type != CMDQ_CREATE_QP_TYPE_GSI)
 			rc = bnxt_re_build_qp1_shadow_qp_recv(qp, wr, &wqe,
 							      payload_sz);
 		if (!rc)
@@ -2622,7 +2625,7 @@ struct ib_cq *bnxt_re_create_cq(struct ib_device *ibdev,
 			goto fail;
 		}
 
-		cq->umem = ib_umem_get(context, req.cq_va,
+		cq->umem = ib_umem_get(udata, req.cq_va,
 				       entries * sizeof(struct cq_base),
 				       IB_ACCESS_LOCAL_WRITE, 1);
 		if (IS_ERR(cq->umem)) {
@@ -3122,19 +3125,33 @@ static void bnxt_re_process_res_shadow_qp_wc(struct bnxt_re_qp *qp,
 	}
 }
 
-static void bnxt_re_process_res_ud_wc(struct ib_wc *wc,
+static void bnxt_re_process_res_ud_wc(struct bnxt_re_qp *qp,
+				      struct ib_wc *wc,
 				      struct bnxt_qplib_cqe *cqe)
 {
+	u8 nw_type;
+
 	wc->opcode = IB_WC_RECV;
 	wc->status = __rc_to_ib_wc_status(cqe->status);
 
-	if (cqe->flags & CQ_RES_RC_FLAGS_IMM)
+	if (cqe->flags & CQ_RES_UD_FLAGS_IMM)
 		wc->wc_flags |= IB_WC_WITH_IMM;
-	if (cqe->flags & CQ_RES_RC_FLAGS_INV)
-		wc->wc_flags |= IB_WC_WITH_INVALIDATE;
-	if ((cqe->flags & (CQ_RES_RC_FLAGS_RDMA | CQ_RES_RC_FLAGS_IMM)) ==
-	    (CQ_RES_RC_FLAGS_RDMA | CQ_RES_RC_FLAGS_IMM))
-		wc->opcode = IB_WC_RECV_RDMA_WITH_IMM;
+	/* report only on GSI QP for Thor */
+	if (qp->qplib_qp.type == CMDQ_CREATE_QP_TYPE_GSI) {
+		wc->wc_flags |= IB_WC_GRH;
+		memcpy(wc->smac, cqe->smac, ETH_ALEN);
+		wc->wc_flags |= IB_WC_WITH_SMAC;
+		if (cqe->flags & CQ_RES_UD_FLAGS_META_FORMAT_VLAN) {
+			wc->vlan_id = (cqe->cfa_meta & 0xFFF);
+			if (wc->vlan_id < 0x1000)
+				wc->wc_flags |= IB_WC_WITH_VLAN;
+		}
+		nw_type = (cqe->flags & CQ_RES_UD_FLAGS_ROCE_IP_VER_MASK) >>
+			   CQ_RES_UD_FLAGS_ROCE_IP_VER_SFT;
+		wc->network_hdr_type = bnxt_re_to_ib_nw_type(nw_type);
+		wc->wc_flags |= IB_WC_WITH_NETWORK_HDR_TYPE;
+	}
+
 }
 
 static int send_phantom_wqe(struct bnxt_re_qp *qp)
@@ -3226,7 +3243,7 @@ int bnxt_re_poll_cq(struct ib_cq *ib_cq, int num_entries, struct ib_wc *wc)
 
 			switch (cqe->opcode) {
 			case CQ_BASE_CQE_TYPE_REQ:
-				if (qp->qplib_qp.id ==
+				if (qp->rdev->qp1_sqp && qp->qplib_qp.id ==
 				    qp->rdev->qp1_sqp->qplib_qp.id) {
 					/* Handle this completion with
 					 * the stored completion
@@ -3261,7 +3278,7 @@ int bnxt_re_poll_cq(struct ib_cq *ib_cq, int num_entries, struct ib_wc *wc)
 				bnxt_re_process_res_rc_wc(wc, cqe);
 				break;
 			case CQ_BASE_CQE_TYPE_RES_UD:
-				if (qp->qplib_qp.id ==
+				if (qp->rdev->qp1_sqp && qp->qplib_qp.id ==
 				    qp->rdev->qp1_sqp->qplib_qp.id) {
 					/* Handle this completion with
 					 * the stored completion
@@ -3274,7 +3291,7 @@ int bnxt_re_poll_cq(struct ib_cq *ib_cq, int num_entries, struct ib_wc *wc)
 						break;
 					}
 				}
-				bnxt_re_process_res_ud_wc(wc, cqe);
+				bnxt_re_process_res_ud_wc(qp, wc, cqe);
 				break;
 			default:
 				dev_err(rdev_to_dev(cq->rdev),
@@ -3301,10 +3318,10 @@ int bnxt_re_req_notify_cq(struct ib_cq *ib_cq,
 	spin_lock_irqsave(&cq->cq_lock, flags);
 	/* Trigger on the very next completion */
 	if (ib_cqn_flags & IB_CQ_NEXT_COMP)
-		type = DBR_DBR_TYPE_CQ_ARMALL;
+		type = DBC_DBC_TYPE_CQ_ARMALL;
 	/* Trigger on the next solicited completion */
 	else if (ib_cqn_flags & IB_CQ_SOLICITED)
-		type = DBR_DBR_TYPE_CQ_ARMSE;
+		type = DBC_DBC_TYPE_CQ_ARMSE;
 
 	/* Poll to see if there are missed events */
 	if ((ib_cqn_flags & IB_CQ_REPORT_MISSED_EVENTS) &&
@@ -3537,19 +3554,14 @@ static int fill_umem_pbl_tbl(struct ib_umem *umem, u64 *pbl_tbl_orig,
 	u64 *pbl_tbl = pbl_tbl_orig;
 	u64 paddr;
 	u64 page_mask = (1ULL << page_shift) - 1;
-	int i, pages;
-	struct scatterlist *sg;
-	int entry;
-
-	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {
-		pages = sg_dma_len(sg) >> PAGE_SHIFT;
-		for (i = 0; i < pages; i++) {
-			paddr = sg_dma_address(sg) + (i << PAGE_SHIFT);
-			if (pbl_tbl == pbl_tbl_orig)
-				*pbl_tbl++ = paddr & ~page_mask;
-			else if ((paddr & page_mask) == 0)
-				*pbl_tbl++ = paddr;
-		}
+	struct sg_dma_page_iter sg_iter;
+
+	for_each_sg_dma_page (umem->sg_head.sgl, &sg_iter, umem->nmap, 0) {
+		paddr = sg_page_iter_dma_address(&sg_iter);
+		if (pbl_tbl == pbl_tbl_orig)
+			*pbl_tbl++ = paddr & ~page_mask;
+		else if ((paddr & page_mask) == 0)
+			*pbl_tbl++ = paddr;
 	}
 	return pbl_tbl - pbl_tbl_orig;
 }
@@ -3589,8 +3601,7 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length,
 	/* The fixed portion of the rkey is the same as the lkey */
 	mr->ib_mr.rkey = mr->qplib_mr.rkey;
 
-	umem = ib_umem_get(ib_pd->uobject->context, start, length,
-			   mr_access_flags, 0);
+	umem = ib_umem_get(udata, start, length, mr_access_flags, 0);
 	if (IS_ERR(umem)) {
 		dev_err(rdev_to_dev(rdev), "Failed to get umem");
 		rc = -EFAULT;
@@ -3613,7 +3624,7 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length,
 		goto free_umem;
 	}
 
-	page_shift = umem->page_shift;
+	page_shift = PAGE_SHIFT;
 
 	if (!bnxt_re_page_size_ok(page_shift)) {
 		dev_err(rdev_to_dev(rdev), "umem page size unsupported!");
@@ -3660,13 +3671,15 @@ free_mr:
 	return ERR_PTR(rc);
 }
 
-struct ib_ucontext *bnxt_re_alloc_ucontext(struct ib_device *ibdev,
-					   struct ib_udata *udata)
+int bnxt_re_alloc_ucontext(struct ib_ucontext *ctx, struct ib_udata *udata)
 {
+	struct ib_device *ibdev = ctx->device;
+	struct bnxt_re_ucontext *uctx =
+		container_of(ctx, struct bnxt_re_ucontext, ib_uctx);
 	struct bnxt_re_dev *rdev = to_bnxt_re_dev(ibdev, ibdev);
-	struct bnxt_re_uctx_resp resp;
-	struct bnxt_re_ucontext *uctx;
 	struct bnxt_qplib_dev_attr *dev_attr = &rdev->dev_attr;
+	struct bnxt_re_uctx_resp resp;
+	u32 chip_met_rev_num = 0;
 	int rc;
 
 	dev_dbg(rdev_to_dev(rdev), "ABI version requested %d",
@@ -3675,13 +3688,9 @@ struct ib_ucontext *bnxt_re_alloc_ucontext(struct ib_device *ibdev,
 	if (ibdev->uverbs_abi_ver != BNXT_RE_ABI_VERSION) {
 		dev_dbg(rdev_to_dev(rdev), " is different from the device %d ",
 			BNXT_RE_ABI_VERSION);
-		return ERR_PTR(-EPERM);
+		return -EPERM;
 	}
 
-	uctx = kzalloc(sizeof(*uctx), GFP_KERNEL);
-	if (!uctx)
-		return ERR_PTR(-ENOMEM);
-
 	uctx->rdev = rdev;
 
 	uctx->shpg = (void *)__get_free_page(GFP_KERNEL);
@@ -3691,37 +3700,45 @@ struct ib_ucontext *bnxt_re_alloc_ucontext(struct ib_device *ibdev,
 	}
 	spin_lock_init(&uctx->sh_lock);
 
-	resp.dev_id = rdev->en_dev->pdev->devfn; /*Temp, Use idr_alloc instead*/
+	resp.comp_mask = BNXT_RE_UCNTX_CMASK_HAVE_CCTX;
+	chip_met_rev_num = rdev->chip_ctx.chip_num;
+	chip_met_rev_num |= ((u32)rdev->chip_ctx.chip_rev & 0xFF) <<
+			     BNXT_RE_CHIP_ID0_CHIP_REV_SFT;
+	chip_met_rev_num |= ((u32)rdev->chip_ctx.chip_metal & 0xFF) <<
+			     BNXT_RE_CHIP_ID0_CHIP_MET_SFT;
+	resp.chip_id0 = chip_met_rev_num;
+	/* Future extension of chip info */
+	resp.chip_id1 = 0;
+	/*Temp, Use idr_alloc instead */
+	resp.dev_id = rdev->en_dev->pdev->devfn;
 	resp.max_qp = rdev->qplib_ctx.qpc_count;
 	resp.pg_size = PAGE_SIZE;
 	resp.cqe_sz = sizeof(struct cq_base);
 	resp.max_cqd = dev_attr->max_cq_wqes;
 	resp.rsvd    = 0;
 
-	rc = ib_copy_to_udata(udata, &resp, sizeof(resp));
+	rc = ib_copy_to_udata(udata, &resp, min(udata->outlen, sizeof(resp)));
 	if (rc) {
 		dev_err(rdev_to_dev(rdev), "Failed to copy user context");
 		rc = -EFAULT;
 		goto cfail;
 	}
 
-	return &uctx->ib_uctx;
+	return 0;
 cfail:
 	free_page((unsigned long)uctx->shpg);
 	uctx->shpg = NULL;
 fail:
-	kfree(uctx);
-	return ERR_PTR(rc);
+	return rc;
 }
 
-int bnxt_re_dealloc_ucontext(struct ib_ucontext *ib_uctx)
+void bnxt_re_dealloc_ucontext(struct ib_ucontext *ib_uctx)
 {
 	struct bnxt_re_ucontext *uctx = container_of(ib_uctx,
 						   struct bnxt_re_ucontext,
 						   ib_uctx);
 
 	struct bnxt_re_dev *rdev = uctx->rdev;
-	int rc = 0;
 
 	if (uctx->shpg)
 		free_page((unsigned long)uctx->shpg);
@@ -3730,17 +3747,10 @@ int bnxt_re_dealloc_ucontext(struct ib_ucontext *ib_uctx)
 		/* Free DPI only if this is the first PD allocated by the
 		 * application and mark the context dpi as NULL
 		 */
-		rc = bnxt_qplib_dealloc_dpi(&rdev->qplib_res,
-					    &rdev->qplib_res.dpi_tbl,
-					    &uctx->dpi);
-		if (rc)
-			dev_err(rdev_to_dev(rdev), "Deallocate HW DPI failed!");
-			/* Don't fail, continue*/
+		bnxt_qplib_dealloc_dpi(&rdev->qplib_res,
+				       &rdev->qplib_res.dpi_tbl, &uctx->dpi);
 		uctx->dpi.dbr = NULL;
 	}
-
-	kfree(uctx);
-	return 0;
 }
 
 /* Helper function to mmap the virtual memory from user app */
diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.h b/drivers/infiniband/hw/bnxt_re/ib_verbs.h
index c4af72604b4f..e45465ed4eee 100644
--- a/drivers/infiniband/hw/bnxt_re/ib_verbs.h
+++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.h
@@ -56,8 +56,8 @@ struct bnxt_re_fence_data {
 };
 
 struct bnxt_re_pd {
+	struct ib_pd            ib_pd;
 	struct bnxt_re_dev	*rdev;
-	struct ib_pd		ib_pd;
 	struct bnxt_qplib_pd	qplib_pd;
 	struct bnxt_re_fence_data fence;
 };
@@ -135,8 +135,8 @@ struct bnxt_re_mw {
 };
 
 struct bnxt_re_ucontext {
+	struct ib_ucontext      ib_uctx;
 	struct bnxt_re_dev	*rdev;
-	struct ib_ucontext	ib_uctx;
 	struct bnxt_qplib_dpi	dpi;
 	void			*shpg;
 	spinlock_t		sh_lock;	/* protect shpg */
@@ -163,10 +163,9 @@ int bnxt_re_query_gid(struct ib_device *ibdev, u8 port_num,
 		      int index, union ib_gid *gid);
 enum rdma_link_layer bnxt_re_get_link_layer(struct ib_device *ibdev,
 					    u8 port_num);
-struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev,
-			       struct ib_ucontext *context,
-			       struct ib_udata *udata);
-int bnxt_re_dealloc_pd(struct ib_pd *pd);
+int bnxt_re_alloc_pd(struct ib_pd *pd, struct ib_ucontext *context,
+		     struct ib_udata *udata);
+void bnxt_re_dealloc_pd(struct ib_pd *pd);
 struct ib_ah *bnxt_re_create_ah(struct ib_pd *pd,
 				struct rdma_ah_attr *ah_attr,
 				u32 flags,
@@ -216,9 +215,8 @@ int bnxt_re_dealloc_mw(struct ib_mw *mw);
 struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 				  u64 virt_addr, int mr_access_flags,
 				  struct ib_udata *udata);
-struct ib_ucontext *bnxt_re_alloc_ucontext(struct ib_device *ibdev,
-					   struct ib_udata *udata);
-int bnxt_re_dealloc_ucontext(struct ib_ucontext *context);
+int bnxt_re_alloc_ucontext(struct ib_ucontext *ctx, struct ib_udata *udata);
+void bnxt_re_dealloc_ucontext(struct ib_ucontext *context);
 int bnxt_re_mmap(struct ib_ucontext *context, struct vm_area_struct *vma);
 
 unsigned long bnxt_re_lock_cqs(struct bnxt_re_qp *qp);
diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c
index e7a997f2a537..2bd24ac45ee4 100644
--- a/drivers/infiniband/hw/bnxt_re/main.c
+++ b/drivers/infiniband/hw/bnxt_re/main.c
@@ -80,6 +80,29 @@ static DEFINE_MUTEX(bnxt_re_dev_lock);
 static struct workqueue_struct *bnxt_re_wq;
 static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev);
 
+static void bnxt_re_destroy_chip_ctx(struct bnxt_re_dev *rdev)
+{
+	rdev->rcfw.res = NULL;
+	rdev->qplib_res.cctx = NULL;
+}
+
+static int bnxt_re_setup_chip_ctx(struct bnxt_re_dev *rdev)
+{
+	struct bnxt_en_dev *en_dev;
+	struct bnxt *bp;
+
+	en_dev = rdev->en_dev;
+	bp = netdev_priv(en_dev->net);
+
+	rdev->chip_ctx.chip_num = bp->chip_num;
+	/* rest members to follow eventually */
+
+	rdev->qplib_res.cctx = &rdev->chip_ctx;
+	rdev->rcfw.res = &rdev->qplib_res;
+
+	return 0;
+}
+
 /* SR-IOV helper functions */
 
 static void bnxt_re_get_sriov_func_type(struct bnxt_re_dev *rdev)
@@ -278,6 +301,7 @@ static int bnxt_re_register_netdev(struct bnxt_re_dev *rdev)
 
 	rc = en_dev->en_ops->bnxt_register_device(en_dev, BNXT_ROCE_ULP,
 						  &bnxt_re_ulp_ops, rdev);
+	rdev->qplib_res.pdev = rdev->en_dev->pdev;
 	return rc;
 }
 
@@ -345,7 +369,8 @@ static void bnxt_re_fill_fw_msg(struct bnxt_fw_msg *fw_msg, void *msg,
 	fw_msg->timeout = timeout;
 }
 
-static int bnxt_re_net_ring_free(struct bnxt_re_dev *rdev, u16 fw_ring_id)
+static int bnxt_re_net_ring_free(struct bnxt_re_dev *rdev,
+				 u16 fw_ring_id, int type)
 {
 	struct bnxt_en_dev *en_dev = rdev->en_dev;
 	struct hwrm_ring_free_input req = {0};
@@ -359,7 +384,7 @@ static int bnxt_re_net_ring_free(struct bnxt_re_dev *rdev, u16 fw_ring_id)
 	memset(&fw_msg, 0, sizeof(fw_msg));
 
 	bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_RING_FREE, -1, -1);
-	req.ring_type = RING_ALLOC_REQ_RING_TYPE_L2_CMPL;
+	req.ring_type = type;
 	req.ring_id = cpu_to_le16(fw_ring_id);
 	bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp,
 			    sizeof(resp), DFLT_HWRM_CMD_TIMEOUT);
@@ -396,7 +421,7 @@ static int bnxt_re_net_ring_alloc(struct bnxt_re_dev *rdev, dma_addr_t *dma_arr,
 	/* Association of ring index with doorbell index and MSIX number */
 	req.logical_id = cpu_to_le16(map_index);
 	req.length = cpu_to_le32(ring_mask + 1);
-	req.ring_type = RING_ALLOC_REQ_RING_TYPE_L2_CMPL;
+	req.ring_type = type;
 	req.int_mode = RING_ALLOC_REQ_INT_MODE_MSIX;
 	bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp,
 			    sizeof(resp), DFLT_HWRM_CMD_TIMEOUT);
@@ -538,7 +563,8 @@ static struct bnxt_en_dev *bnxt_re_dev_probe(struct net_device *netdev)
 static ssize_t hw_rev_show(struct device *device, struct device_attribute *attr,
 			   char *buf)
 {
-	struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev);
+	struct bnxt_re_dev *rdev =
+		rdma_device_to_drv_device(device, struct bnxt_re_dev, ibdev);
 
 	return scnprintf(buf, PAGE_SIZE, "0x%x\n", rdev->en_dev->pdev->vendor);
 }
@@ -547,7 +573,8 @@ static DEVICE_ATTR_RO(hw_rev);
 static ssize_t hca_type_show(struct device *device,
 			     struct device_attribute *attr, char *buf)
 {
-	struct bnxt_re_dev *rdev = to_bnxt_re_dev(device, ibdev.dev);
+	struct bnxt_re_dev *rdev =
+		rdma_device_to_drv_device(device, struct bnxt_re_dev, ibdev);
 
 	return scnprintf(buf, PAGE_SIZE, "%s\n", rdev->ibdev.node_desc);
 }
@@ -610,6 +637,8 @@ static const struct ib_device_ops bnxt_re_dev_ops = {
 	.query_srq = bnxt_re_query_srq,
 	.reg_user_mr = bnxt_re_reg_user_mr,
 	.req_notify_cq = bnxt_re_req_notify_cq,
+	INIT_RDMA_OBJ_SIZE(ib_pd, bnxt_re_pd, ib_pd),
+	INIT_RDMA_OBJ_SIZE(ib_ucontext, bnxt_re_ucontext, ib_uctx),
 };
 
 static int bnxt_re_register_ib(struct bnxt_re_dev *rdev)
@@ -662,7 +691,7 @@ static int bnxt_re_register_ib(struct bnxt_re_dev *rdev)
 	rdma_set_device_sysfs_group(ibdev, &bnxt_re_dev_attr_group);
 	ibdev->driver_id = RDMA_DRIVER_BNXT_RE;
 	ib_set_device_ops(ibdev, &bnxt_re_dev_ops);
-	return ib_register_device(ibdev, "bnxt_re%d", NULL);
+	return ib_register_device(ibdev, "bnxt_re%d");
 }
 
 static void bnxt_re_dev_remove(struct bnxt_re_dev *rdev)
@@ -686,7 +715,7 @@ static struct bnxt_re_dev *bnxt_re_dev_add(struct net_device *netdev,
 	struct bnxt_re_dev *rdev;
 
 	/* Allocate bnxt_re_dev instance here */
-	rdev = (struct bnxt_re_dev *)ib_alloc_device(sizeof(*rdev));
+	rdev = ib_alloc_device(bnxt_re_dev, ibdev);
 	if (!rdev) {
 		dev_err(NULL, "%s: bnxt_re_dev allocation failure!",
 			ROCE_DRV_MODULE_NAME);
@@ -858,6 +887,12 @@ static int bnxt_re_cqn_handler(struct bnxt_qplib_nq *nq,
 	return 0;
 }
 
+static u32 bnxt_re_get_nqdb_offset(struct bnxt_re_dev *rdev, u16 indx)
+{
+	return bnxt_qplib_is_chip_gen_p5(&rdev->chip_ctx) ?
+				0x10000 : rdev->msix_entries[indx].db_offset;
+}
+
 static void bnxt_re_cleanup_res(struct bnxt_re_dev *rdev)
 {
 	int i;
@@ -871,18 +906,18 @@ static void bnxt_re_cleanup_res(struct bnxt_re_dev *rdev)
 
 static int bnxt_re_init_res(struct bnxt_re_dev *rdev)
 {
-	int rc = 0, i;
 	int num_vec_enabled = 0;
+	int rc = 0, i;
+	u32 db_offt;
 
 	bnxt_qplib_init_res(&rdev->qplib_res);
 
 	for (i = 1; i < rdev->num_msix ; i++) {
+		db_offt = bnxt_re_get_nqdb_offset(rdev, i);
 		rc = bnxt_qplib_enable_nq(rdev->en_dev->pdev, &rdev->nq[i - 1],
 					  i - 1, rdev->msix_entries[i].vector,
-					  rdev->msix_entries[i].db_offset,
-					  &bnxt_re_cqn_handler,
+					  db_offt, &bnxt_re_cqn_handler,
 					  &bnxt_re_srqn_handler);
-
 		if (rc) {
 			dev_err(rdev_to_dev(rdev),
 				"Failed to enable NQ with rc = 0x%x", rc);
@@ -894,16 +929,18 @@ static int bnxt_re_init_res(struct bnxt_re_dev *rdev)
 fail:
 	for (i = num_vec_enabled; i >= 0; i--)
 		bnxt_qplib_disable_nq(&rdev->nq[i]);
-
 	return rc;
 }
 
 static void bnxt_re_free_nq_res(struct bnxt_re_dev *rdev)
 {
+	u8 type;
 	int i;
 
 	for (i = 0; i < rdev->num_msix - 1; i++) {
-		bnxt_re_net_ring_free(rdev, rdev->nq[i].ring_id);
+		type = bnxt_qplib_get_ring_type(&rdev->chip_ctx);
+		bnxt_re_net_ring_free(rdev, rdev->nq[i].ring_id, type);
+		rdev->nq[i].res = NULL;
 		bnxt_qplib_free_nq(&rdev->nq[i]);
 	}
 }
@@ -925,8 +962,11 @@ static void bnxt_re_free_res(struct bnxt_re_dev *rdev)
 
 static int bnxt_re_alloc_res(struct bnxt_re_dev *rdev)
 {
-	int rc = 0, i;
 	int num_vec_created = 0;
+	dma_addr_t *pg_map;
+	int rc = 0, i;
+	int pages;
+	u8 type;
 
 	/* Configure and allocate resources for qplib */
 	rdev->qplib_res.rcfw = &rdev->rcfw;
@@ -947,6 +987,7 @@ static int bnxt_re_alloc_res(struct bnxt_re_dev *rdev)
 		goto dealloc_res;
 
 	for (i = 0; i < rdev->num_msix - 1; i++) {
+		rdev->nq[i].res = &rdev->qplib_res;
 		rdev->nq[i].hwq.max_elements = BNXT_RE_MAX_CQ_COUNT +
 			BNXT_RE_MAX_SRQC_COUNT + 2;
 		rc = bnxt_qplib_alloc_nq(rdev->en_dev->pdev, &rdev->nq[i]);
@@ -955,13 +996,13 @@ static int bnxt_re_alloc_res(struct bnxt_re_dev *rdev)
 				i, rc);
 			goto free_nq;
 		}
-		rc = bnxt_re_net_ring_alloc
-			(rdev, rdev->nq[i].hwq.pbl[PBL_LVL_0].pg_map_arr,
-			 rdev->nq[i].hwq.pbl[rdev->nq[i].hwq.level].pg_count,
-			 HWRM_RING_ALLOC_CMPL,
-			 BNXT_QPLIB_NQE_MAX_CNT - 1,
-			 rdev->msix_entries[i + 1].ring_idx,
-			 &rdev->nq[i].ring_id);
+		type = bnxt_qplib_get_ring_type(&rdev->chip_ctx);
+		pg_map = rdev->nq[i].hwq.pbl[PBL_LVL_0].pg_map_arr;
+		pages = rdev->nq[i].hwq.pbl[rdev->nq[i].hwq.level].pg_count;
+		rc = bnxt_re_net_ring_alloc(rdev, pg_map, pages, type,
+					    BNXT_QPLIB_NQE_MAX_CNT - 1,
+					    rdev->msix_entries[i + 1].ring_idx,
+					    &rdev->nq[i].ring_id);
 		if (rc) {
 			dev_err(rdev_to_dev(rdev),
 				"Failed to allocate NQ fw id with rc = 0x%x",
@@ -974,7 +1015,8 @@ static int bnxt_re_alloc_res(struct bnxt_re_dev *rdev)
 	return 0;
 free_nq:
 	for (i = num_vec_created; i >= 0; i--) {
-		bnxt_re_net_ring_free(rdev, rdev->nq[i].ring_id);
+		type = bnxt_qplib_get_ring_type(&rdev->chip_ctx);
+		bnxt_re_net_ring_free(rdev, rdev->nq[i].ring_id, type);
 		bnxt_qplib_free_nq(&rdev->nq[i]);
 	}
 	bnxt_qplib_dealloc_dpi(&rdev->qplib_res,
@@ -1228,6 +1270,7 @@ static void bnxt_re_query_hwrm_intf_version(struct bnxt_re_dev *rdev)
 
 static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev)
 {
+	u8 type;
 	int rc;
 
 	if (test_and_clear_bit(BNXT_RE_FLAG_IBDEV_REGISTERED, &rdev->flags)) {
@@ -1251,7 +1294,8 @@ static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev)
 		bnxt_re_net_stats_ctx_free(rdev, rdev->qplib_ctx.stats.fw_id);
 		bnxt_qplib_free_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx);
 		bnxt_qplib_disable_rcfw_channel(&rdev->rcfw);
-		bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id);
+		type = bnxt_qplib_get_ring_type(&rdev->chip_ctx);
+		bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id, type);
 		bnxt_qplib_free_rcfw_channel(&rdev->rcfw);
 	}
 	if (test_and_clear_bit(BNXT_RE_FLAG_GOT_MSIX, &rdev->flags)) {
@@ -1260,6 +1304,8 @@ static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev)
 			dev_warn(rdev_to_dev(rdev),
 				 "Failed to free MSI-X vectors: %#x", rc);
 	}
+
+	bnxt_re_destroy_chip_ctx(rdev);
 	if (test_and_clear_bit(BNXT_RE_FLAG_NETDEV_REGISTERED, &rdev->flags)) {
 		rc = bnxt_re_unregister_netdev(rdev);
 		if (rc)
@@ -1280,9 +1326,12 @@ static void bnxt_re_worker(struct work_struct *work)
 
 static int bnxt_re_ib_reg(struct bnxt_re_dev *rdev)
 {
-	int rc;
-
+	dma_addr_t *pg_map;
+	u32 db_offt, ridx;
+	int pages, vid;
 	bool locked;
+	u8 type;
+	int rc;
 
 	/* Acquire rtnl lock through out this function */
 	rtnl_lock();
@@ -1297,6 +1346,12 @@ static int bnxt_re_ib_reg(struct bnxt_re_dev *rdev)
 	}
 	set_bit(BNXT_RE_FLAG_NETDEV_REGISTERED, &rdev->flags);
 
+	rc = bnxt_re_setup_chip_ctx(rdev);
+	if (rc) {
+		dev_err(rdev_to_dev(rdev), "Failed to get chip context\n");
+		return -EINVAL;
+	}
+
 	/* Check whether VF or PF */
 	bnxt_re_get_sriov_func_type(rdev);
 
@@ -1320,21 +1375,22 @@ static int bnxt_re_ib_reg(struct bnxt_re_dev *rdev)
 		pr_err("Failed to allocate RCFW Channel: %#x\n", rc);
 		goto fail;
 	}
-	rc = bnxt_re_net_ring_alloc
-			(rdev, rdev->rcfw.creq.pbl[PBL_LVL_0].pg_map_arr,
-			 rdev->rcfw.creq.pbl[rdev->rcfw.creq.level].pg_count,
-			 HWRM_RING_ALLOC_CMPL, BNXT_QPLIB_CREQE_MAX_CNT - 1,
-			 rdev->msix_entries[BNXT_RE_AEQ_IDX].ring_idx,
-			 &rdev->rcfw.creq_ring_id);
+	type = bnxt_qplib_get_ring_type(&rdev->chip_ctx);
+	pg_map = rdev->rcfw.creq.pbl[PBL_LVL_0].pg_map_arr;
+	pages = rdev->rcfw.creq.pbl[rdev->rcfw.creq.level].pg_count;
+	ridx = rdev->msix_entries[BNXT_RE_AEQ_IDX].ring_idx;
+	rc = bnxt_re_net_ring_alloc(rdev, pg_map, pages, type,
+				    BNXT_QPLIB_CREQE_MAX_CNT - 1,
+				    ridx, &rdev->rcfw.creq_ring_id);
 	if (rc) {
 		pr_err("Failed to allocate CREQ: %#x\n", rc);
 		goto free_rcfw;
 	}
-	rc = bnxt_qplib_enable_rcfw_channel
-				(rdev->en_dev->pdev, &rdev->rcfw,
-				 rdev->msix_entries[BNXT_RE_AEQ_IDX].vector,
-				 rdev->msix_entries[BNXT_RE_AEQ_IDX].db_offset,
-				 rdev->is_virtfn, &bnxt_re_aeq_handler);
+	db_offt = bnxt_re_get_nqdb_offset(rdev, BNXT_RE_AEQ_IDX);
+	vid = rdev->msix_entries[BNXT_RE_AEQ_IDX].vector;
+	rc = bnxt_qplib_enable_rcfw_channel(rdev->en_dev->pdev, &rdev->rcfw,
+					    vid, db_offt, rdev->is_virtfn,
+					    &bnxt_re_aeq_handler);
 	if (rc) {
 		pr_err("Failed to enable RCFW channel: %#x\n", rc);
 		goto free_ring;
@@ -1347,7 +1403,8 @@ static int bnxt_re_ib_reg(struct bnxt_re_dev *rdev)
 	if (!rdev->is_virtfn)
 		bnxt_re_set_resource_limits(rdev);
 
-	rc = bnxt_qplib_alloc_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx, 0);
+	rc = bnxt_qplib_alloc_ctx(rdev->en_dev->pdev, &rdev->qplib_ctx, 0,
+				  bnxt_qplib_is_chip_gen_p5(&rdev->chip_ctx));
 	if (rc) {
 		pr_err("Failed to allocate QPLIB context: %#x\n", rc);
 		goto disable_rcfw;
@@ -1418,7 +1475,8 @@ free_ctx:
 disable_rcfw:
 	bnxt_qplib_disable_rcfw_channel(&rdev->rcfw);
 free_ring:
-	bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id);
+	type = bnxt_qplib_get_ring_type(&rdev->chip_ctx);
+	bnxt_re_net_ring_free(rdev, rdev->rcfw.creq_ring_id, type);
 free_rcfw:
 	bnxt_qplib_free_rcfw_channel(&rdev->rcfw);
 fail:
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
index b98b054148cd..71c34d5b0ac0 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c
@@ -44,6 +44,7 @@
 #include <linux/slab.h>
 #include <linux/pci.h>
 #include <linux/prefetch.h>
+#include <linux/if_ether.h>
 
 #include "roce_hsi.h"
 
@@ -244,6 +245,7 @@ static void bnxt_qplib_service_nq(unsigned long data)
 	u16 type;
 	int budget = nq->budget;
 	uintptr_t q_handle;
+	bool gen_p5 = bnxt_qplib_is_chip_gen_p5(nq->res->cctx);
 
 	/* Service the NQ until empty */
 	raw_cons = hwq->cons;
@@ -290,7 +292,7 @@ static void bnxt_qplib_service_nq(unsigned long data)
 			q_handle |= (u64)le32_to_cpu(nqsrqe->srq_handle_high)
 				     << 32;
 			bnxt_qplib_arm_srq((struct bnxt_qplib_srq *)q_handle,
-					   DBR_DBR_TYPE_SRQ_ARMENA);
+					   DBC_DBC_TYPE_SRQ_ARMENA);
 			if (!nq->srqn_handler(nq,
 					      (struct bnxt_qplib_srq *)q_handle,
 					      nqsrqe->event))
@@ -312,7 +314,9 @@ static void bnxt_qplib_service_nq(unsigned long data)
 	}
 	if (hwq->cons != raw_cons) {
 		hwq->cons = raw_cons;
-		NQ_DB_REARM(nq->bar_reg_iomem, hwq->cons, hwq->max_elements);
+		bnxt_qplib_ring_nq_db_rearm(nq->bar_reg_iomem, hwq->cons,
+					    hwq->max_elements, nq->ring_id,
+					    gen_p5);
 	}
 }
 
@@ -336,9 +340,11 @@ static irqreturn_t bnxt_qplib_nq_irq(int irq, void *dev_instance)
 
 void bnxt_qplib_nq_stop_irq(struct bnxt_qplib_nq *nq, bool kill)
 {
+	bool gen_p5 = bnxt_qplib_is_chip_gen_p5(nq->res->cctx);
 	tasklet_disable(&nq->worker);
 	/* Mask h/w interrupt */
-	NQ_DB(nq->bar_reg_iomem, nq->hwq.cons, nq->hwq.max_elements);
+	bnxt_qplib_ring_nq_db(nq->bar_reg_iomem, nq->hwq.cons,
+			      nq->hwq.max_elements, nq->ring_id, gen_p5);
 	/* Sync with last running IRQ handler */
 	synchronize_irq(nq->vector);
 	if (kill)
@@ -373,6 +379,7 @@ void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq)
 int bnxt_qplib_nq_start_irq(struct bnxt_qplib_nq *nq, int nq_indx,
 			    int msix_vector, bool need_init)
 {
+	bool gen_p5 = bnxt_qplib_is_chip_gen_p5(nq->res->cctx);
 	int rc;
 
 	if (nq->requested)
@@ -399,7 +406,8 @@ int bnxt_qplib_nq_start_irq(struct bnxt_qplib_nq *nq, int nq_indx,
 			 nq->vector, nq_indx);
 	}
 	nq->requested = true;
-	NQ_DB_REARM(nq->bar_reg_iomem, nq->hwq.cons, nq->hwq.max_elements);
+	bnxt_qplib_ring_nq_db_rearm(nq->bar_reg_iomem, nq->hwq.cons,
+				    nq->hwq.max_elements, nq->ring_id, gen_p5);
 
 	return rc;
 }
@@ -433,7 +441,8 @@ int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq,
 		rc = -ENOMEM;
 		goto fail;
 	}
-	nq->bar_reg_iomem = ioremap_nocache(nq_base + nq->bar_reg_off, 4);
+	/* Unconditionally map 8 bytes to support 57500 series */
+	nq->bar_reg_iomem = ioremap_nocache(nq_base + nq->bar_reg_off, 8);
 	if (!nq->bar_reg_iomem) {
 		rc = -ENOMEM;
 		goto fail;
@@ -462,15 +471,17 @@ void bnxt_qplib_free_nq(struct bnxt_qplib_nq *nq)
 
 int bnxt_qplib_alloc_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq)
 {
+	u8 hwq_type;
+
 	nq->pdev = pdev;
 	if (!nq->hwq.max_elements ||
 	    nq->hwq.max_elements > BNXT_QPLIB_NQE_MAX_CNT)
 		nq->hwq.max_elements = BNXT_QPLIB_NQE_MAX_CNT;
-
+	hwq_type = bnxt_qplib_get_hwq_type(nq->res);
 	if (bnxt_qplib_alloc_init_hwq(nq->pdev, &nq->hwq, NULL, 0,
 				      &nq->hwq.max_elements,
 				      BNXT_QPLIB_MAX_NQE_ENTRY_SIZE, 0,
-				      PAGE_SIZE, HWQ_TYPE_L2_CMPL))
+				      PAGE_SIZE, hwq_type))
 		return -ENOMEM;
 
 	nq->budget = 8;
@@ -481,21 +492,19 @@ int bnxt_qplib_alloc_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq)
 static void bnxt_qplib_arm_srq(struct bnxt_qplib_srq *srq, u32 arm_type)
 {
 	struct bnxt_qplib_hwq *srq_hwq = &srq->hwq;
-	struct dbr_dbr db_msg = { 0 };
 	void __iomem *db;
-	u32 sw_prod = 0;
+	u32 sw_prod;
+	u64 val = 0;
 
 	/* Ring DB */
-	sw_prod = (arm_type == DBR_DBR_TYPE_SRQ_ARM) ? srq->threshold :
-		   HWQ_CMP(srq_hwq->prod, srq_hwq);
-	db_msg.index = cpu_to_le32((sw_prod << DBR_DBR_INDEX_SFT) &
-				   DBR_DBR_INDEX_MASK);
-	db_msg.type_xid = cpu_to_le32(((srq->id << DBR_DBR_XID_SFT) &
-					DBR_DBR_XID_MASK) | arm_type);
-	db = (arm_type == DBR_DBR_TYPE_SRQ_ARMENA) ?
-		srq->dbr_base : srq->dpi->dbr;
-	wmb(); /* barrier before db ring */
-	__iowrite64_copy(db, &db_msg, sizeof(db_msg) / sizeof(u64));
+	sw_prod = (arm_type == DBC_DBC_TYPE_SRQ_ARM) ?
+		   srq->threshold : HWQ_CMP(srq_hwq->prod, srq_hwq);
+	db = (arm_type == DBC_DBC_TYPE_SRQ_ARMENA) ? srq->dbr_base :
+						     srq->dpi->dbr;
+	val = ((srq->id << DBC_DBC_XID_SFT) & DBC_DBC_XID_MASK) | arm_type;
+	val <<= 32;
+	val |= (sw_prod << DBC_DBC_INDEX_SFT) & DBC_DBC_INDEX_MASK;
+	writeq(val, db);
 }
 
 int bnxt_qplib_destroy_srq(struct bnxt_qplib_res *res,
@@ -590,7 +599,7 @@ int bnxt_qplib_create_srq(struct bnxt_qplib_res *res,
 	srq->id = le32_to_cpu(resp.xid);
 	srq->dbr_base = res->dpi_tbl.dbr_bar_reg_iomem;
 	if (srq->threshold)
-		bnxt_qplib_arm_srq(srq, DBR_DBR_TYPE_SRQ_ARMENA);
+		bnxt_qplib_arm_srq(srq, DBC_DBC_TYPE_SRQ_ARMENA);
 	srq->arm_req = false;
 
 	return 0;
@@ -614,7 +623,7 @@ int bnxt_qplib_modify_srq(struct bnxt_qplib_res *res,
 				    srq_hwq->max_elements - sw_cons + sw_prod;
 	if (count > srq->threshold) {
 		srq->arm_req = false;
-		bnxt_qplib_arm_srq(srq, DBR_DBR_TYPE_SRQ_ARM);
+		bnxt_qplib_arm_srq(srq, DBC_DBC_TYPE_SRQ_ARM);
 	} else {
 		/* Deferred arming */
 		srq->arm_req = true;
@@ -702,10 +711,10 @@ int bnxt_qplib_post_srq_recv(struct bnxt_qplib_srq *srq,
 				    srq_hwq->max_elements - sw_cons + sw_prod;
 	spin_unlock(&srq_hwq->lock);
 	/* Ring DB */
-	bnxt_qplib_arm_srq(srq, DBR_DBR_TYPE_SRQ);
+	bnxt_qplib_arm_srq(srq, DBC_DBC_TYPE_SRQ);
 	if (srq->arm_req == true && count > srq->threshold) {
 		srq->arm_req = false;
-		bnxt_qplib_arm_srq(srq, DBR_DBR_TYPE_SRQ_ARM);
+		bnxt_qplib_arm_srq(srq, DBC_DBC_TYPE_SRQ_ARM);
 	}
 done:
 	return rc;
@@ -853,18 +862,19 @@ exit:
 int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
 {
 	struct bnxt_qplib_rcfw *rcfw = res->rcfw;
-	struct sq_send *hw_sq_send_hdr, **hw_sq_send_ptr;
-	struct cmdq_create_qp req;
-	struct creq_create_qp_resp resp;
-	struct bnxt_qplib_pbl *pbl;
-	struct sq_psn_search **psn_search_ptr;
 	unsigned long int psn_search, poff = 0;
+	struct sq_psn_search **psn_search_ptr;
 	struct bnxt_qplib_q *sq = &qp->sq;
 	struct bnxt_qplib_q *rq = &qp->rq;
+	int i, rc, req_size, psn_sz = 0;
+	struct sq_send **hw_sq_send_ptr;
+	struct creq_create_qp_resp resp;
 	struct bnxt_qplib_hwq *xrrq;
-	int i, rc, req_size, psn_sz;
 	u16 cmd_flags = 0, max_ssge;
-	u32 sw_prod, qp_flags = 0;
+	struct cmdq_create_qp req;
+	struct bnxt_qplib_pbl *pbl;
+	u32 qp_flags = 0;
+	u16 max_rsge;
 
 	RCFW_CMD_PREP(req, CREATE_QP, cmd_flags);
 
@@ -874,8 +884,11 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
 	req.qp_handle = cpu_to_le64(qp->qp_handle);
 
 	/* SQ */
-	psn_sz = (qp->type == CMDQ_CREATE_QP_TYPE_RC) ?
-		 sizeof(struct sq_psn_search) : 0;
+	if (qp->type == CMDQ_CREATE_QP_TYPE_RC) {
+		psn_sz = bnxt_qplib_is_chip_gen_p5(res->cctx) ?
+			 sizeof(struct sq_psn_search_ext) :
+			 sizeof(struct sq_psn_search);
+	}
 	sq->hwq.max_elements = sq->max_wqe;
 	rc = bnxt_qplib_alloc_init_hwq(res->pdev, &sq->hwq, sq->sglist,
 				       sq->nmap, &sq->hwq.max_elements,
@@ -905,10 +918,16 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
 			poff = (psn_search & ~PAGE_MASK) /
 				BNXT_QPLIB_MAX_PSNE_ENTRY_SIZE;
 		}
-		for (i = 0; i < sq->hwq.max_elements; i++)
+		for (i = 0; i < sq->hwq.max_elements; i++) {
 			sq->swq[i].psn_search =
 				&psn_search_ptr[get_psne_pg(i + poff)]
 					       [get_psne_idx(i + poff)];
+			/*psns_ext will be used only for P5 chips. */
+			sq->swq[i].psn_ext =
+				(struct sq_psn_search_ext *)
+				&psn_search_ptr[get_psne_pg(i + poff)]
+					       [get_psne_idx(i + poff)];
+		}
 	}
 	pbl = &sq->hwq.pbl[PBL_LVL_0];
 	req.sq_pbl = cpu_to_le64(pbl->pg_map_arr[0]);
@@ -929,14 +948,6 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
 				CMDQ_CREATE_QP_SQ_PG_SIZE_PG_1G :
 		 CMDQ_CREATE_QP_SQ_PG_SIZE_PG_4K);
 
-	/* initialize all SQ WQEs to LOCAL_INVALID (sq prep for hw fetch) */
-	hw_sq_send_ptr = (struct sq_send **)sq->hwq.pbl_ptr;
-	for (sw_prod = 0; sw_prod < sq->hwq.max_elements; sw_prod++) {
-		hw_sq_send_hdr = &hw_sq_send_ptr[get_sqe_pg(sw_prod)]
-						[get_sqe_idx(sw_prod)];
-		hw_sq_send_hdr->wqe_type = SQ_BASE_WQE_TYPE_LOCAL_INVALID;
-	}
-
 	if (qp->scq)
 		req.scq_cid = cpu_to_le32(qp->scq->id);
 
@@ -1007,8 +1018,9 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
 	req.sq_fwo_sq_sge = cpu_to_le16(
 				((max_ssge & CMDQ_CREATE_QP_SQ_SGE_MASK)
 				 << CMDQ_CREATE_QP_SQ_SGE_SFT) | 0);
+	max_rsge = bnxt_qplib_is_chip_gen_p5(res->cctx) ? 6 : rq->max_sge;
 	req.rq_fwo_rq_sge = cpu_to_le16(
-				((rq->max_sge & CMDQ_CREATE_QP_RQ_SGE_MASK)
+				((max_rsge & CMDQ_CREATE_QP_RQ_SGE_MASK)
 				 << CMDQ_CREATE_QP_RQ_SGE_SFT) | 0);
 	/* ORRQ and IRRQ */
 	if (psn_sz) {
@@ -1053,6 +1065,7 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp)
 
 	qp->id = le32_to_cpu(resp.xid);
 	qp->cur_qp_state = CMDQ_MODIFY_QP_NEW_STATE_RESET;
+	qp->cctx = res->cctx;
 	INIT_LIST_HEAD(&qp->sq_flush);
 	INIT_LIST_HEAD(&qp->rq_flush);
 	rcfw->qp_tbl[qp->id].qp_id = qp->id;
@@ -1494,19 +1507,16 @@ void *bnxt_qplib_get_qp1_rq_buf(struct bnxt_qplib_qp *qp,
 void bnxt_qplib_post_send_db(struct bnxt_qplib_qp *qp)
 {
 	struct bnxt_qplib_q *sq = &qp->sq;
-	struct dbr_dbr db_msg = { 0 };
 	u32 sw_prod;
+	u64 val = 0;
 
+	val = (((qp->id << DBC_DBC_XID_SFT) & DBC_DBC_XID_MASK) |
+	       DBC_DBC_TYPE_SQ);
+	val <<= 32;
 	sw_prod = HWQ_CMP(sq->hwq.prod, &sq->hwq);
-
-	db_msg.index = cpu_to_le32((sw_prod << DBR_DBR_INDEX_SFT) &
-				   DBR_DBR_INDEX_MASK);
-	db_msg.type_xid =
-		cpu_to_le32(((qp->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) |
-			    DBR_DBR_TYPE_SQ);
+	val |= (sw_prod << DBC_DBC_INDEX_SFT) & DBC_DBC_INDEX_MASK;
 	/* Flush all the WQE writes to HW */
-	wmb();
-	__iowrite64_copy(qp->dpi->dbr, &db_msg, sizeof(db_msg) / sizeof(u64));
+	writeq(val, qp->dpi->dbr);
 }
 
 int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp,
@@ -1617,7 +1627,8 @@ int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp,
 				((offsetof(typeof(*sqe), data) + 15) >> 4);
 		sqe->inv_key_or_imm_data = cpu_to_le32(
 						wqe->send.inv_key);
-		if (qp->type == CMDQ_CREATE_QP_TYPE_UD) {
+		if (qp->type == CMDQ_CREATE_QP_TYPE_UD ||
+		    qp->type == CMDQ_CREATE_QP_TYPE_GSI) {
 			sqe->q_key = cpu_to_le32(wqe->send.q_key);
 			sqe->dst_qp = cpu_to_le32(
 					wqe->send.dst_qp & SQ_SEND_DST_QP_MASK);
@@ -1741,14 +1752,26 @@ int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp,
 	}
 	swq->next_psn = sq->psn & BTH_PSN_MASK;
 	if (swq->psn_search) {
-		swq->psn_search->opcode_start_psn = cpu_to_le32(
-			((swq->start_psn << SQ_PSN_SEARCH_START_PSN_SFT) &
-			 SQ_PSN_SEARCH_START_PSN_MASK) |
-			((wqe->type << SQ_PSN_SEARCH_OPCODE_SFT) &
-			 SQ_PSN_SEARCH_OPCODE_MASK));
-		swq->psn_search->flags_next_psn = cpu_to_le32(
-			((swq->next_psn << SQ_PSN_SEARCH_NEXT_PSN_SFT) &
-			 SQ_PSN_SEARCH_NEXT_PSN_MASK));
+		u32 opcd_spsn;
+		u32 flg_npsn;
+
+		opcd_spsn = ((swq->start_psn << SQ_PSN_SEARCH_START_PSN_SFT) &
+			      SQ_PSN_SEARCH_START_PSN_MASK);
+		opcd_spsn |= ((wqe->type << SQ_PSN_SEARCH_OPCODE_SFT) &
+			       SQ_PSN_SEARCH_OPCODE_MASK);
+		flg_npsn = ((swq->next_psn << SQ_PSN_SEARCH_NEXT_PSN_SFT) &
+			     SQ_PSN_SEARCH_NEXT_PSN_MASK);
+		if (bnxt_qplib_is_chip_gen_p5(qp->cctx)) {
+			swq->psn_ext->opcode_start_psn =
+						cpu_to_le32(opcd_spsn);
+			swq->psn_ext->flags_next_psn =
+						cpu_to_le32(flg_npsn);
+		} else {
+			swq->psn_search->opcode_start_psn =
+						cpu_to_le32(opcd_spsn);
+			swq->psn_search->flags_next_psn =
+						cpu_to_le32(flg_npsn);
+		}
 	}
 queue_err:
 	if (sch_handler) {
@@ -1785,19 +1808,16 @@ done:
 void bnxt_qplib_post_recv_db(struct bnxt_qplib_qp *qp)
 {
 	struct bnxt_qplib_q *rq = &qp->rq;
-	struct dbr_dbr db_msg = { 0 };
 	u32 sw_prod;
+	u64 val = 0;
 
+	val = (((qp->id << DBC_DBC_XID_SFT) & DBC_DBC_XID_MASK) |
+	       DBC_DBC_TYPE_RQ);
+	val <<= 32;
 	sw_prod = HWQ_CMP(rq->hwq.prod, &rq->hwq);
-	db_msg.index = cpu_to_le32((sw_prod << DBR_DBR_INDEX_SFT) &
-				   DBR_DBR_INDEX_MASK);
-	db_msg.type_xid =
-		cpu_to_le32(((qp->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) |
-			    DBR_DBR_TYPE_RQ);
-
+	val |= (sw_prod << DBC_DBC_INDEX_SFT) & DBC_DBC_INDEX_MASK;
 	/* Flush the writes to HW Rx WQE before the ringing Rx DB */
-	wmb();
-	__iowrite64_copy(qp->dpi->dbr, &db_msg, sizeof(db_msg) / sizeof(u64));
+	writeq(val, qp->dpi->dbr);
 }
 
 int bnxt_qplib_post_recv(struct bnxt_qplib_qp *qp,
@@ -1881,32 +1901,28 @@ done:
 /* Spinlock must be held */
 static void bnxt_qplib_arm_cq_enable(struct bnxt_qplib_cq *cq)
 {
-	struct dbr_dbr db_msg = { 0 };
+	u64 val = 0;
 
-	db_msg.type_xid =
-		cpu_to_le32(((cq->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) |
-			    DBR_DBR_TYPE_CQ_ARMENA);
+	val = ((cq->id << DBC_DBC_XID_SFT) & DBC_DBC_XID_MASK) |
+	       DBC_DBC_TYPE_CQ_ARMENA;
+	val <<= 32;
 	/* Flush memory writes before enabling the CQ */
-	wmb();
-	__iowrite64_copy(cq->dbr_base, &db_msg, sizeof(db_msg) / sizeof(u64));
+	writeq(val, cq->dbr_base);
 }
 
 static void bnxt_qplib_arm_cq(struct bnxt_qplib_cq *cq, u32 arm_type)
 {
 	struct bnxt_qplib_hwq *cq_hwq = &cq->hwq;
-	struct dbr_dbr db_msg = { 0 };
 	u32 sw_cons;
+	u64 val = 0;
 
 	/* Ring DB */
+	val = ((cq->id << DBC_DBC_XID_SFT) & DBC_DBC_XID_MASK) | arm_type;
+	val <<= 32;
 	sw_cons = HWQ_CMP(cq_hwq->cons, cq_hwq);
-	db_msg.index = cpu_to_le32((sw_cons << DBR_DBR_INDEX_SFT) &
-				    DBR_DBR_INDEX_MASK);
-	db_msg.type_xid =
-		cpu_to_le32(((cq->id << DBR_DBR_XID_SFT) & DBR_DBR_XID_MASK) |
-			    arm_type);
+	val |= (sw_cons << DBC_DBC_INDEX_SFT) & DBC_DBC_INDEX_MASK;
 	/* flush memory writes before arming the CQ */
-	wmb();
-	__iowrite64_copy(cq->dpi->dbr, &db_msg, sizeof(db_msg) / sizeof(u64));
+	writeq(val, cq->dpi->dbr);
 }
 
 int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq)
@@ -2053,6 +2069,7 @@ static int __flush_rq(struct bnxt_qplib_q *rq, struct bnxt_qplib_qp *qp,
 		opcode = CQ_BASE_CQE_TYPE_RES_RC;
 		break;
 	case CMDQ_CREATE_QP_TYPE_UD:
+	case CMDQ_CREATE_QP_TYPE_GSI:
 		opcode = CQ_BASE_CQE_TYPE_RES_UD;
 		break;
 	}
@@ -2125,7 +2142,7 @@ static int do_wa9060(struct bnxt_qplib_qp *qp, struct bnxt_qplib_cq *cq,
 		sq->send_phantom = true;
 
 		/* TODO: Only ARM if the previous SQE is ARMALL */
-		bnxt_qplib_arm_cq(cq, DBR_DBR_TYPE_CQ_ARMALL);
+		bnxt_qplib_arm_cq(cq, DBC_DBC_TYPE_CQ_ARMALL);
 
 		rc = -EAGAIN;
 		goto out;
@@ -2410,12 +2427,14 @@ static int bnxt_qplib_cq_process_res_ud(struct bnxt_qplib_cq *cq,
 	}
 	cqe = *pcqe;
 	cqe->opcode = hwcqe->cqe_type_toggle & CQ_BASE_CQE_TYPE_MASK;
-	cqe->length = le32_to_cpu(hwcqe->length);
+	cqe->length = (u32)le16_to_cpu(hwcqe->length);
+	cqe->cfa_meta = le16_to_cpu(hwcqe->cfa_metadata);
 	cqe->invrkey = le32_to_cpu(hwcqe->imm_data);
 	cqe->flags = le16_to_cpu(hwcqe->flags);
 	cqe->status = hwcqe->status;
 	cqe->qp_handle = (u64)(unsigned long)qp;
-	memcpy(cqe->smac, hwcqe->src_mac, 6);
+	/*FIXME: Endianness fix needed for smace */
+	memcpy(cqe->smac, hwcqe->src_mac, ETH_ALEN);
 	wr_id_idx = le32_to_cpu(hwcqe->src_qp_high_srq_or_rq_wr_id)
 				& CQ_RES_UD_SRQ_OR_RQ_WR_ID_MASK;
 	cqe->src_qp = le16_to_cpu(hwcqe->src_qp_low) |
@@ -2794,7 +2813,7 @@ int bnxt_qplib_poll_cq(struct bnxt_qplib_cq *cq, struct bnxt_qplib_cqe *cqe,
 	}
 	if (cq->hwq.cons != raw_cons) {
 		cq->hwq.cons = raw_cons;
-		bnxt_qplib_arm_cq(cq, DBR_DBR_TYPE_CQ);
+		bnxt_qplib_arm_cq(cq, DBC_DBC_TYPE_CQ);
 	}
 exit:
 	return num_cqes - budget;
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.h b/drivers/infiniband/hw/bnxt_re/qplib_fp.h
index 72352ca80ace..3f618b5f1f06 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_fp.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.h
@@ -106,6 +106,7 @@ struct bnxt_qplib_swq {
 	u32				start_psn;
 	u32				next_psn;
 	struct sq_psn_search		*psn_search;
+	struct sq_psn_search_ext	*psn_ext;
 };
 
 struct bnxt_qplib_swqe {
@@ -254,6 +255,7 @@ struct bnxt_qplib_q {
 struct bnxt_qplib_qp {
 	struct bnxt_qplib_pd		*pd;
 	struct bnxt_qplib_dpi		*dpi;
+	struct bnxt_qplib_chip_ctx	*cctx;
 	u64				qp_handle;
 #define        BNXT_QPLIB_QP_ID_INVALID        0xFFFFFFFF
 	u32				id;
@@ -347,6 +349,7 @@ struct bnxt_qplib_cqe {
 	u8				type;
 	u8				opcode;
 	u32				length;
+	u16				cfa_meta;
 	u64				wr_id;
 	union {
 		__be32			immdata;
@@ -432,13 +435,47 @@ struct bnxt_qplib_cq {
 #define NQ_DB_CP_FLAGS			(NQ_DB_KEY_CP    |	\
 					 NQ_DB_IDX_VALID |	\
 					 NQ_DB_IRQ_DIS)
-#define NQ_DB_REARM(db, raw_cons, cp_bit)			\
-	writel(NQ_DB_CP_FLAGS_REARM | ((raw_cons) & ((cp_bit) - 1)), db)
-#define NQ_DB(db, raw_cons, cp_bit)				\
-	writel(NQ_DB_CP_FLAGS | ((raw_cons) & ((cp_bit) - 1)), db)
+
+static inline void bnxt_qplib_ring_nq_db64(void __iomem *db, u32 index,
+					   u32 xid, bool arm)
+{
+	u64 val;
+
+	val = xid & DBC_DBC_XID_MASK;
+	val |= DBC_DBC_PATH_ROCE;
+	val |= arm ? DBC_DBC_TYPE_NQ_ARM : DBC_DBC_TYPE_NQ;
+	val <<= 32;
+	val |= index & DBC_DBC_INDEX_MASK;
+	writeq(val, db);
+}
+
+static inline void bnxt_qplib_ring_nq_db_rearm(void __iomem *db, u32 raw_cons,
+					       u32 max_elements, u32 xid,
+					       bool gen_p5)
+{
+	u32 index = raw_cons & (max_elements - 1);
+
+	if (gen_p5)
+		bnxt_qplib_ring_nq_db64(db, index, xid, true);
+	else
+		writel(NQ_DB_CP_FLAGS_REARM | (index & DBC_DBC32_XID_MASK), db);
+}
+
+static inline void bnxt_qplib_ring_nq_db(void __iomem *db, u32 raw_cons,
+					 u32 max_elements, u32 xid,
+					 bool gen_p5)
+{
+	u32 index = raw_cons & (max_elements - 1);
+
+	if (gen_p5)
+		bnxt_qplib_ring_nq_db64(db, index, xid, false);
+	else
+		writel(NQ_DB_CP_FLAGS | (index & DBC_DBC32_XID_MASK), db);
+}
 
 struct bnxt_qplib_nq {
 	struct pci_dev		*pdev;
+	struct bnxt_qplib_res	*res;
 
 	int			vector;
 	cpumask_t		mask;
@@ -448,7 +485,7 @@ struct bnxt_qplib_nq {
 	struct bnxt_qplib_hwq	hwq;
 
 	u16			bar_reg;
-	u16			bar_reg_off;
+	u32			bar_reg_off;
 	u16			ring_id;
 	void __iomem		*bar_reg_iomem;
 
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
index 19551aa43850..c6461e957078 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c
@@ -359,11 +359,12 @@ static int bnxt_qplib_process_qp_event(struct bnxt_qplib_rcfw *rcfw,
 static void bnxt_qplib_service_creq(unsigned long data)
 {
 	struct bnxt_qplib_rcfw *rcfw = (struct bnxt_qplib_rcfw *)data;
+	bool gen_p5 = bnxt_qplib_is_chip_gen_p5(rcfw->res->cctx);
 	struct bnxt_qplib_hwq *creq = &rcfw->creq;
+	u32 type, budget = CREQ_ENTRY_POLL_BUDGET;
 	struct creq_base *creqe, **creq_ptr;
 	u32 sw_cons, raw_cons;
 	unsigned long flags;
-	u32 type, budget = CREQ_ENTRY_POLL_BUDGET;
 
 	/* Service the CREQ until budget is over */
 	spin_lock_irqsave(&creq->lock, flags);
@@ -407,8 +408,9 @@ static void bnxt_qplib_service_creq(unsigned long data)
 
 	if (creq->cons != raw_cons) {
 		creq->cons = raw_cons;
-		CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, raw_cons,
-			      creq->max_elements);
+		bnxt_qplib_ring_creq_db_rearm(rcfw->creq_bar_reg_iomem,
+					      raw_cons, creq->max_elements,
+					      rcfw->creq_ring_id, gen_p5);
 	}
 	spin_unlock_irqrestore(&creq->lock, flags);
 }
@@ -480,11 +482,13 @@ int bnxt_qplib_init_rcfw(struct bnxt_qplib_rcfw *rcfw,
 	req.log2_dbr_pg_size = cpu_to_le16(PAGE_SHIFT -
 					   RCFW_DBR_BASE_PAGE_SHIFT);
 	/*
-	 * VFs need not setup the HW context area, PF
+	 * Gen P5 devices doesn't require this allocation
+	 * as the L2 driver does the same for RoCE also.
+	 * Also, VFs need not setup the HW context area, PF
 	 * shall setup this area for VF. Skipping the
 	 * HW programming
 	 */
-	if (is_virtfn)
+	if (is_virtfn || bnxt_qplib_is_chip_gen_p5(rcfw->res->cctx))
 		goto skip_ctx_setup;
 
 	level = ctx->qpc_tbl.level;
@@ -560,12 +564,15 @@ int bnxt_qplib_alloc_rcfw_channel(struct pci_dev *pdev,
 				  struct bnxt_qplib_ctx *ctx,
 				  int qp_tbl_sz)
 {
+	u8 hwq_type;
+
 	rcfw->pdev = pdev;
 	rcfw->creq.max_elements = BNXT_QPLIB_CREQE_MAX_CNT;
+	hwq_type = bnxt_qplib_get_hwq_type(rcfw->res);
 	if (bnxt_qplib_alloc_init_hwq(rcfw->pdev, &rcfw->creq, NULL, 0,
 				      &rcfw->creq.max_elements,
-				      BNXT_QPLIB_CREQE_UNITS, 0, PAGE_SIZE,
-				      HWQ_TYPE_L2_CMPL)) {
+				      BNXT_QPLIB_CREQE_UNITS,
+				      0, PAGE_SIZE, hwq_type)) {
 		dev_err(&rcfw->pdev->dev,
 			"HW channel CREQ allocation failed\n");
 		goto fail;
@@ -607,10 +614,13 @@ fail:
 
 void bnxt_qplib_rcfw_stop_irq(struct bnxt_qplib_rcfw *rcfw, bool kill)
 {
+	bool gen_p5 = bnxt_qplib_is_chip_gen_p5(rcfw->res->cctx);
+
 	tasklet_disable(&rcfw->worker);
 	/* Mask h/w interrupts */
-	CREQ_DB(rcfw->creq_bar_reg_iomem, rcfw->creq.cons,
-		rcfw->creq.max_elements);
+	bnxt_qplib_ring_creq_db(rcfw->creq_bar_reg_iomem, rcfw->creq.cons,
+				rcfw->creq.max_elements, rcfw->creq_ring_id,
+				gen_p5);
 	/* Sync with last running IRQ-handler */
 	synchronize_irq(rcfw->vector);
 	if (kill)
@@ -647,6 +657,7 @@ void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw)
 int bnxt_qplib_rcfw_start_irq(struct bnxt_qplib_rcfw *rcfw, int msix_vector,
 			      bool need_init)
 {
+	bool gen_p5 = bnxt_qplib_is_chip_gen_p5(rcfw->res->cctx);
 	int rc;
 
 	if (rcfw->requested)
@@ -663,8 +674,9 @@ int bnxt_qplib_rcfw_start_irq(struct bnxt_qplib_rcfw *rcfw, int msix_vector,
 	if (rc)
 		return rc;
 	rcfw->requested = true;
-	CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, rcfw->creq.cons,
-		      rcfw->creq.max_elements);
+	bnxt_qplib_ring_creq_db_rearm(rcfw->creq_bar_reg_iomem,
+				      rcfw->creq.cons, rcfw->creq.max_elements,
+				      rcfw->creq_ring_id, gen_p5);
 
 	return 0;
 }
@@ -684,8 +696,7 @@ int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev,
 	/* General */
 	rcfw->seq_num = 0;
 	set_bit(FIRMWARE_FIRST_FLAG, &rcfw->flags);
-	bmap_size = BITS_TO_LONGS(rcfw->cmdq_depth *
-				  sizeof(unsigned long));
+	bmap_size = BITS_TO_LONGS(rcfw->cmdq_depth) * sizeof(unsigned long);
 	rcfw->cmdq_bitmap = kzalloc(bmap_size, GFP_KERNEL);
 	if (!rcfw->cmdq_bitmap)
 		return -ENOMEM;
@@ -718,8 +729,9 @@ int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev,
 		dev_err(&rcfw->pdev->dev,
 			"CREQ BAR region %d resc start is 0!\n",
 			rcfw->creq_bar_reg);
+	/* Unconditionally map 8 bytes to support 57500 series */
 	rcfw->creq_bar_reg_iomem = ioremap_nocache(res_base + cp_bar_reg_off,
-						   4);
+						   8);
 	if (!rcfw->creq_bar_reg_iomem) {
 		dev_err(&rcfw->pdev->dev, "CREQ BAR region %d mapping failed\n",
 			rcfw->creq_bar_reg);
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
index be0ef0e8c53e..2138533bb642 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h
@@ -157,10 +157,46 @@ static inline u32 get_creq_idx(u32 val)
 #define CREQ_DB_CP_FLAGS		(CREQ_DB_KEY_CP |	\
 					 CREQ_DB_IDX_VALID |	\
 					 CREQ_DB_IRQ_DIS)
-#define CREQ_DB_REARM(db, raw_cons, cp_bit)			\
-	writel(CREQ_DB_CP_FLAGS_REARM | ((raw_cons) & ((cp_bit) - 1)), db)
-#define CREQ_DB(db, raw_cons, cp_bit)				\
-	writel(CREQ_DB_CP_FLAGS | ((raw_cons) & ((cp_bit) - 1)), db)
+
+static inline void bnxt_qplib_ring_creq_db64(void __iomem *db, u32 index,
+					     u32 xid, bool arm)
+{
+	u64 val = 0;
+
+	val = xid & DBC_DBC_XID_MASK;
+	val |= DBC_DBC_PATH_ROCE;
+	val |= arm ? DBC_DBC_TYPE_NQ_ARM : DBC_DBC_TYPE_NQ;
+	val <<= 32;
+	val |= index & DBC_DBC_INDEX_MASK;
+
+	writeq(val, db);
+}
+
+static inline void bnxt_qplib_ring_creq_db_rearm(void __iomem *db, u32 raw_cons,
+						 u32 max_elements, u32 xid,
+						 bool gen_p5)
+{
+	u32 index = raw_cons & (max_elements - 1);
+
+	if (gen_p5)
+		bnxt_qplib_ring_creq_db64(db, index, xid, true);
+	else
+		writel(CREQ_DB_CP_FLAGS_REARM | (index & DBC_DBC32_XID_MASK),
+		       db);
+}
+
+static inline void bnxt_qplib_ring_creq_db(void __iomem *db, u32 raw_cons,
+					   u32 max_elements, u32 xid,
+					   bool gen_p5)
+{
+	u32 index = raw_cons & (max_elements - 1);
+
+	if (gen_p5)
+		bnxt_qplib_ring_creq_db64(db, index, xid, true);
+	else
+		writel(CREQ_DB_CP_FLAGS | (index & DBC_DBC32_XID_MASK),
+		       db);
+}
 
 #define CREQ_ENTRY_POLL_BUDGET		0x100
 
@@ -187,6 +223,7 @@ struct bnxt_qplib_qp_node {
 /* RCFW Communication Channels */
 struct bnxt_qplib_rcfw {
 	struct pci_dev		*pdev;
+	struct bnxt_qplib_res	*res;
 	int			vector;
 	struct tasklet_struct	worker;
 	bool			requested;
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.c b/drivers/infiniband/hw/bnxt_re/qplib_res.c
index 57d4951679cb..0bc24f934829 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_res.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_res.c
@@ -85,7 +85,7 @@ static void __free_pbl(struct pci_dev *pdev, struct bnxt_qplib_pbl *pbl,
 static int __alloc_pbl(struct pci_dev *pdev, struct bnxt_qplib_pbl *pbl,
 		       struct scatterlist *sghead, u32 pages, u32 pg_size)
 {
-	struct scatterlist *sg;
+	struct sg_dma_page_iter sg_iter;
 	bool is_umem = false;
 	int i;
 
@@ -116,13 +116,11 @@ static int __alloc_pbl(struct pci_dev *pdev, struct bnxt_qplib_pbl *pbl,
 	} else {
 		i = 0;
 		is_umem = true;
-		for_each_sg(sghead, sg, pages, i) {
-			pbl->pg_map_arr[i] = sg_dma_address(sg);
-			pbl->pg_arr[i] = sg_virt(sg);
-			if (!pbl->pg_arr[i])
-				goto fail;
-
+		for_each_sg_dma_page (sghead, &sg_iter, pages, 0) {
+			pbl->pg_map_arr[i] = sg_page_iter_dma_address(&sg_iter);
+			pbl->pg_arr[i] = NULL;
 			pbl->pg_count++;
+			i++;
 		}
 	}
 
@@ -330,13 +328,13 @@ void bnxt_qplib_free_ctx(struct pci_dev *pdev,
  */
 int bnxt_qplib_alloc_ctx(struct pci_dev *pdev,
 			 struct bnxt_qplib_ctx *ctx,
-			 bool virt_fn)
+			 bool virt_fn, bool is_p5)
 {
 	int i, j, k, rc = 0;
 	int fnz_idx = -1;
 	__le64 **pbl_ptr;
 
-	if (virt_fn)
+	if (virt_fn || is_p5)
 		goto stats_alloc;
 
 	/* QPC Tables */
@@ -762,7 +760,11 @@ static int bnxt_qplib_alloc_stats_ctx(struct pci_dev *pdev,
 {
 	memset(stats, 0, sizeof(*stats));
 	stats->fw_id = -1;
-	stats->size = sizeof(struct ctx_hw_stats);
+	/* 128 byte aligned context memory is required only for 57500.
+	 * However making this unconditional, it does not harm previous
+	 * generation.
+	 */
+	stats->size = ALIGN(sizeof(struct ctx_hw_stats), 128);
 	stats->dma = dma_alloc_coherent(&pdev->dev, stats->size,
 					&stats->dma_map, GFP_KERNEL);
 	if (!stats->dma) {
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_res.h b/drivers/infiniband/hw/bnxt_re/qplib_res.h
index 1e80aa7bbcce..32cebd0f1436 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_res.h
+++ b/drivers/infiniband/hw/bnxt_re/qplib_res.h
@@ -180,12 +180,20 @@ struct bnxt_qplib_ctx {
 	u64				hwrm_intf_ver;
 };
 
+struct bnxt_qplib_chip_ctx {
+	u16	chip_num;
+	u8	chip_rev;
+	u8	chip_metal;
+};
+
+#define CHIP_NUM_57500          0x1750
+
 struct bnxt_qplib_res {
 	struct pci_dev			*pdev;
+	struct bnxt_qplib_chip_ctx	*cctx;
 	struct net_device		*netdev;
 
 	struct bnxt_qplib_rcfw		*rcfw;
-
 	struct bnxt_qplib_pd_tbl	pd_tbl;
 	struct bnxt_qplib_sgid_tbl	sgid_tbl;
 	struct bnxt_qplib_pkey_tbl	pkey_tbl;
@@ -193,6 +201,24 @@ struct bnxt_qplib_res {
 	bool				prio;
 };
 
+static inline bool bnxt_qplib_is_chip_gen_p5(struct bnxt_qplib_chip_ctx *cctx)
+{
+	return (cctx->chip_num == CHIP_NUM_57500);
+}
+
+static inline u8 bnxt_qplib_get_hwq_type(struct bnxt_qplib_res *res)
+{
+	return bnxt_qplib_is_chip_gen_p5(res->cctx) ?
+					HWQ_TYPE_QUEUE : HWQ_TYPE_L2_CMPL;
+}
+
+static inline u8 bnxt_qplib_get_ring_type(struct bnxt_qplib_chip_ctx *cctx)
+{
+	return bnxt_qplib_is_chip_gen_p5(cctx) ?
+	       RING_ALLOC_REQ_RING_TYPE_NQ :
+	       RING_ALLOC_REQ_RING_TYPE_ROCE_CMPL;
+}
+
 #define to_bnxt_qplib(ptr, type, member)	\
 	container_of(ptr, type, member)
 
@@ -226,5 +252,5 @@ void bnxt_qplib_free_ctx(struct pci_dev *pdev,
 			 struct bnxt_qplib_ctx *ctx);
 int bnxt_qplib_alloc_ctx(struct pci_dev *pdev,
 			 struct bnxt_qplib_ctx *ctx,
-			 bool virt_fn);
+			 bool virt_fn, bool is_p5);
 #endif /* __BNXT_QPLIB_RES_H__ */
diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c
index efa0f2949dc7..e9c53e406404 100644
--- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c
+++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c
@@ -119,7 +119,8 @@ int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw,
 	 * reporting the max number
 	 */
 	attr->max_qp_wqes -= BNXT_QPLIB_RESERVED_QP_WRS;
-	attr->max_qp_sges = sb->max_sge;
+	attr->max_qp_sges = bnxt_qplib_is_chip_gen_p5(rcfw->res->cctx) ?
+			    6 : sb->max_sge;
 	attr->max_cq = le32_to_cpu(sb->max_cq);
 	attr->max_cq_wqes = le32_to_cpu(sb->max_cqe);
 	attr->max_cq_sges = attr->max_qp_sges;
diff --git a/drivers/infiniband/hw/bnxt_re/roce_hsi.h b/drivers/infiniband/hw/bnxt_re/roce_hsi.h
index 8a9ead419ac2..e4b09e7c2175 100644
--- a/drivers/infiniband/hw/bnxt_re/roce_hsi.h
+++ b/drivers/infiniband/hw/bnxt_re/roce_hsi.h
@@ -49,11 +49,11 @@ struct cmpl_doorbell {
 	#define CMPL_DOORBELL_IDX_SFT				    0
 	#define CMPL_DOORBELL_RESERVED_MASK			    0x3000000UL
 	#define CMPL_DOORBELL_RESERVED_SFT			    24
-	#define CMPL_DOORBELL_IDX_VALID			    0x4000000UL
+	#define CMPL_DOORBELL_IDX_VALID				    0x4000000UL
 	#define CMPL_DOORBELL_MASK				    0x8000000UL
 	#define CMPL_DOORBELL_KEY_MASK				    0xf0000000UL
 	#define CMPL_DOORBELL_KEY_SFT				    28
-	#define CMPL_DOORBELL_KEY_CMPL				   (0x2UL << 28)
+	#define CMPL_DOORBELL_KEY_CMPL				(0x2UL << 28)
 };
 
 /* Status Door Bell Format (4 bytes) */
@@ -71,46 +71,56 @@ struct status_doorbell {
 /* RoCE Host Structures */
 
 /* Doorbell Structures */
-/* 64b Doorbell Format (8 bytes) */
-struct dbr_dbr {
-	__le32 index;
-	#define DBR_DBR_INDEX_MASK				    0xfffffUL
-	#define DBR_DBR_INDEX_SFT				    0
-	#define DBR_DBR_RESERVED12_MASK			    0xfff00000UL
-	#define DBR_DBR_RESERVED12_SFT				    20
-	__le32 type_xid;
-	#define DBR_DBR_XID_MASK				    0xfffffUL
-	#define DBR_DBR_XID_SFT				    0
-	#define DBR_DBR_RESERVED8_MASK				    0xff00000UL
-	#define DBR_DBR_RESERVED8_SFT				    20
-	#define DBR_DBR_TYPE_MASK				    0xf0000000UL
-	#define DBR_DBR_TYPE_SFT				    28
-	#define DBR_DBR_TYPE_SQ				   (0x0UL << 28)
-	#define DBR_DBR_TYPE_RQ				   (0x1UL << 28)
-	#define DBR_DBR_TYPE_SRQ				   (0x2UL << 28)
-	#define DBR_DBR_TYPE_SRQ_ARM				   (0x3UL << 28)
-	#define DBR_DBR_TYPE_CQ				   (0x4UL << 28)
-	#define DBR_DBR_TYPE_CQ_ARMSE				   (0x5UL << 28)
-	#define DBR_DBR_TYPE_CQ_ARMALL				   (0x6UL << 28)
-	#define DBR_DBR_TYPE_CQ_ARMENA				   (0x7UL << 28)
-	#define DBR_DBR_TYPE_SRQ_ARMENA			   (0x8UL << 28)
-	#define DBR_DBR_TYPE_CQ_CUTOFF_ACK			   (0x9UL << 28)
-	#define DBR_DBR_TYPE_NULL				   (0xfUL << 28)
-};
-
-/* 32b Doorbell Format (4 bytes) */
-struct dbr_dbr32 {
-	__le32 type_abs_incr_xid;
-	#define DBR_DBR32_XID_MASK				    0xfffffUL
-	#define DBR_DBR32_XID_SFT				    0
-	#define DBR_DBR32_RESERVED4_MASK			    0xf00000UL
-	#define DBR_DBR32_RESERVED4_SFT			    20
-	#define DBR_DBR32_INCR_MASK				    0xf000000UL
-	#define DBR_DBR32_INCR_SFT				    24
-	#define DBR_DBR32_ABS					    0x10000000UL
-	#define DBR_DBR32_TYPE_MASK				    0xe0000000UL
-	#define DBR_DBR32_TYPE_SFT				    29
-	#define DBR_DBR32_TYPE_SQ				   (0x0UL << 29)
+/* dbc_dbc (size:64b/8B) */
+struct dbc_dbc {
+	__le32  index;
+	#define DBC_DBC_INDEX_MASK		0xffffffUL
+	#define DBC_DBC_INDEX_SFT		0
+	__le32  type_path_xid;
+	#define DBC_DBC_XID_MASK		0xfffffUL
+	#define DBC_DBC_XID_SFT			0
+	#define DBC_DBC_PATH_MASK		0x3000000UL
+	#define DBC_DBC_PATH_SFT		24
+	#define DBC_DBC_PATH_ROCE		(0x0UL << 24)
+	#define DBC_DBC_PATH_L2			(0x1UL << 24)
+	#define DBC_DBC_PATH_ENGINE		(0x2UL << 24)
+	#define DBC_DBC_PATH_LAST		DBC_DBC_PATH_ENGINE
+	#define DBC_DBC_DEBUG_TRACE		0x8000000UL
+	#define DBC_DBC_TYPE_MASK		0xf0000000UL
+	#define DBC_DBC_TYPE_SFT		28
+	#define DBC_DBC_TYPE_SQ			(0x0UL << 28)
+	#define DBC_DBC_TYPE_RQ			(0x1UL << 28)
+	#define DBC_DBC_TYPE_SRQ		(0x2UL << 28)
+	#define DBC_DBC_TYPE_SRQ_ARM		(0x3UL << 28)
+	#define DBC_DBC_TYPE_CQ			(0x4UL << 28)
+	#define DBC_DBC_TYPE_CQ_ARMSE		(0x5UL << 28)
+	#define DBC_DBC_TYPE_CQ_ARMALL		(0x6UL << 28)
+	#define DBC_DBC_TYPE_CQ_ARMENA		(0x7UL << 28)
+	#define DBC_DBC_TYPE_SRQ_ARMENA		(0x8UL << 28)
+	#define DBC_DBC_TYPE_CQ_CUTOFF_ACK	(0x9UL << 28)
+	#define DBC_DBC_TYPE_NQ			(0xaUL << 28)
+	#define DBC_DBC_TYPE_NQ_ARM		(0xbUL << 28)
+	#define DBC_DBC_TYPE_NULL		(0xfUL << 28)
+	#define DBC_DBC_TYPE_LAST		DBC_DBC_TYPE_NULL
+};
+
+/* dbc_dbc32 (size:32b/4B) */
+struct dbc_dbc32 {
+	__le32  type_abs_incr_xid;
+	#define DBC_DBC32_XID_MASK		0xfffffUL
+	#define DBC_DBC32_XID_SFT		0
+	#define DBC_DBC32_PATH_MASK		0xc00000UL
+	#define DBC_DBC32_PATH_SFT		22
+	#define DBC_DBC32_PATH_ROCE		(0x0UL << 22)
+	#define DBC_DBC32_PATH_L2		(0x1UL << 22)
+	#define DBC_DBC32_PATH_LAST		DBC_DBC32_PATH_L2
+	#define DBC_DBC32_INCR_MASK		0xf000000UL
+	#define DBC_DBC32_INCR_SFT		24
+	#define DBC_DBC32_ABS			0x10000000UL
+	#define DBC_DBC32_TYPE_MASK		0xe0000000UL
+	#define DBC_DBC32_TYPE_SFT		29
+	#define DBC_DBC32_TYPE_SQ		(0x0UL << 29)
+	#define DBC_DBC32_TYPE_LAST		DBC_DBC32_TYPE_SQ
 };
 
 /* SQ WQE Structures */
@@ -149,7 +159,24 @@ struct sq_psn_search {
 	#define SQ_PSN_SEARCH_NEXT_PSN_MASK			    0xffffffUL
 	#define SQ_PSN_SEARCH_NEXT_PSN_SFT			    0
 	#define SQ_PSN_SEARCH_FLAGS_MASK			    0xff000000UL
-	#define SQ_PSN_SEARCH_FLAGS_SFT			    24
+	#define SQ_PSN_SEARCH_FLAGS_SFT				    24
+};
+
+/* sq_psn_search_ext (size:128b/16B) */
+struct sq_psn_search_ext {
+	__le32  opcode_start_psn;
+	#define SQ_PSN_SEARCH_EXT_START_PSN_MASK		    0xffffffUL
+	#define SQ_PSN_SEARCH_EXT_START_PSN_SFT			    0
+	#define SQ_PSN_SEARCH_EXT_OPCODE_MASK			    0xff000000UL
+	#define SQ_PSN_SEARCH_EXT_OPCODE_SFT			    24
+	__le32  flags_next_psn;
+	#define SQ_PSN_SEARCH_EXT_NEXT_PSN_MASK			    0xffffffUL
+	#define SQ_PSN_SEARCH_EXT_NEXT_PSN_SFT			    0
+	#define SQ_PSN_SEARCH_EXT_FLAGS_MASK			    0xff000000UL
+	#define SQ_PSN_SEARCH_EXT_FLAGS_SFT			    24
+	__le16  start_slot_idx;
+	__le16  reserved16;
+	__le32  reserved32;
 };
 
 /* Send SQ WQE (40 bytes) */
@@ -505,22 +532,24 @@ struct cq_res_rc {
 
 /* Responder UD CQE (32 bytes) */
 struct cq_res_ud {
-	__le32 length;
+	__le16 length;
 	#define CQ_RES_UD_LENGTH_MASK				    0x3fffUL
 	#define CQ_RES_UD_LENGTH_SFT				    0
-	#define CQ_RES_UD_RESERVED18_MASK			    0xffffc000UL
-	#define CQ_RES_UD_RESERVED18_SFT			    14
+	__le16 cfa_metadata;
+	#define CQ_RES_UD_CFA_METADATA_VID_MASK			0xfffUL
+	#define CQ_RES_UD_CFA_METADATA_VID_SFT			0
+	#define CQ_RES_UD_CFA_METADATA_DE			0x1000UL
+	#define CQ_RES_UD_CFA_METADATA_PRI_MASK			0xe000UL
+	#define CQ_RES_UD_CFA_METADATA_PRI_SFT			13
 	__le32 imm_data;
 	__le64 qp_handle;
 	__le16 src_mac[3];
 	__le16 src_qp_low;
 	u8 cqe_type_toggle;
-	#define CQ_RES_UD_TOGGLE				    0x1UL
-	#define CQ_RES_UD_CQE_TYPE_MASK			    0x1eUL
-	#define CQ_RES_UD_CQE_TYPE_SFT				    1
+	#define CQ_RES_UD_TOGGLE				   0x1UL
+	#define CQ_RES_UD_CQE_TYPE_MASK				   0x1eUL
+	#define CQ_RES_UD_CQE_TYPE_SFT				   1
 	#define CQ_RES_UD_CQE_TYPE_RES_UD			   (0x2UL << 1)
-	#define CQ_RES_UD_RESERVED3_MASK			    0xe0UL
-	#define CQ_RES_UD_RESERVED3_SFT			    5
 	u8 status;
 	#define CQ_RES_UD_STATUS_OK				   0x0UL
 	#define CQ_RES_UD_STATUS_LOCAL_ACCESS_ERROR		   0x1UL
@@ -536,18 +565,30 @@ struct cq_res_ud {
 	#define CQ_RES_UD_FLAGS_SRQ_SRQ			   (0x1UL << 0)
 	#define CQ_RES_UD_FLAGS_SRQ_LAST    CQ_RES_UD_FLAGS_SRQ_SRQ
 	#define CQ_RES_UD_FLAGS_IMM				    0x2UL
-	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_MASK		    0xcUL
-	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_SFT		    2
-	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_V1			   (0x0UL << 2)
-	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV4		   (0x2UL << 2)
-	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV6		   (0x3UL << 2)
+	#define CQ_RES_UD_FLAGS_UNUSED_MASK			0xcUL
+	#define CQ_RES_UD_FLAGS_UNUSED_SFT			2
+	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_MASK		0x30UL
+	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_SFT			4
+	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_V1			(0x0UL << 4)
+	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV4		(0x2UL << 4)
+	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV6		(0x3UL << 4)
 	#define CQ_RES_UD_FLAGS_ROCE_IP_VER_LAST		\
 					CQ_RES_UD_FLAGS_ROCE_IP_VER_V2IPV6
+	#define CQ_RES_UD_FLAGS_META_FORMAT_MASK		0x3c0UL
+	#define CQ_RES_UD_FLAGS_META_FORMAT_SFT			6
+	#define CQ_RES_UD_FLAGS_META_FORMAT_NONE		(0x0UL << 6)
+	#define CQ_RES_UD_FLAGS_META_FORMAT_VLAN		(0x1UL << 6)
+	#define CQ_RES_UD_FLAGS_META_FORMAT_TUNNEL_ID		(0x2UL << 6)
+	#define CQ_RES_UD_FLAGS_META_FORMAT_CHDR_DATA		(0x3UL << 6)
+	#define CQ_RES_UD_FLAGS_META_FORMAT_HDR_OFFSET		(0x4UL << 6)
+	#define CQ_RES_UD_FLAGS_META_FORMAT_LAST		\
+					CQ_RES_UD_FLAGS_META_FORMAT_HDR_OFFSET
+	#define CQ_RES_UD_FLAGS_EXT_META_FORMAT_MASK		0xc00UL
+	#define CQ_RES_UD_FLAGS_EXT_META_FORMAT_SFT		10
+
 	__le32 src_qp_high_srq_or_rq_wr_id;
 	#define CQ_RES_UD_SRQ_OR_RQ_WR_ID_MASK			    0xfffffUL
 	#define CQ_RES_UD_SRQ_OR_RQ_WR_ID_SFT			    0
-	#define CQ_RES_UD_RESERVED4_MASK			    0xf00000UL
-	#define CQ_RES_UD_RESERVED4_SFT			    20
 	#define CQ_RES_UD_SRC_QP_HIGH_MASK			    0xff000000UL
 	#define CQ_RES_UD_SRC_QP_HIGH_SFT			    24
 };
@@ -983,6 +1024,7 @@ struct cmdq_create_qp {
 	#define CMDQ_CREATE_QP_TYPE_RC				   0x2UL
 	#define CMDQ_CREATE_QP_TYPE_UD				   0x4UL
 	#define CMDQ_CREATE_QP_TYPE_RAW_ETHERTYPE		   0x6UL
+	#define CMDQ_CREATE_QP_TYPE_GSI				   0x7UL
 	u8 sq_pg_size_sq_lvl;
 	#define CMDQ_CREATE_QP_SQ_LVL_MASK			    0xfUL
 	#define CMDQ_CREATE_QP_SQ_LVL_SFT			    0
@@ -2719,6 +2761,8 @@ struct creq_query_func_resp_sb {
 	__le16 max_srq;
 	__le32 max_gid;
 	__le32 tqm_alloc_reqs[12];
+	__le32 max_dpi;
+	__le32 reserved_32;
 };
 
 /* Set resources command response (16 bytes) */