Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:

 1) New offloading infrastructure and example 'rocker' driver for
    offloading of switching and routing to hardware.

    This work was done by a large group of dedicated individuals, not
    limited to: Scott Feldman, Jiri Pirko, Thomas Graf, John Fastabend,
    Jamal Hadi Salim, Andy Gospodarek, Florian Fainelli, Roopa Prabhu

 2) Start making the networking operate on IOV iterators instead of
    modifying iov objects in-situ during transfers.  Thanks to Al Viro
    and Herbert Xu.

 3) A set of new netlink interfaces for the TIPC stack, from Richard
    Alpe.

 4) Remove unnecessary looping during ipv6 routing lookups, from Martin
    KaFai Lau.

 5) Add PAUSE frame generation support to gianfar driver, from Matei
    Pavaluca.

 6) Allow for larger reordering levels in TCP, which are easily
    achievable in the real world right now, from Eric Dumazet.

 7) Add a variable of napi_schedule that doesn't need to disable cpu
    interrupts, from Eric Dumazet.

 8) Use a doubly linked list to optimize neigh_parms_release(), from
    Nicolas Dichtel.

 9) Various enhancements to the kernel BPF verifier, and allow eBPF
    programs to actually be attached to sockets.  From Alexei
    Starovoitov.

10) Support TSO/LSO in sunvnet driver, from David L Stevens.

11) Allow controlling ECN usage via routing metrics, from Florian
    Westphal.

12) Remote checksum offload, from Tom Herbert.

13) Add split-header receive, BQL, and xmit_more support to amd-xgbe
    driver, from Thomas Lendacky.

14) Add MPLS support to openvswitch, from Simon Horman.

15) Support wildcard tunnel endpoints in ipv6 tunnels, from Steffen
    Klassert.

16) Do gro flushes on a per-device basis using a timer, from Eric
    Dumazet.  This tries to resolve the conflicting goals between the
    desired handling of bulk vs.  RPC-like traffic.

17) Allow userspace to ask for the CPU upon what a packet was
    received/steered, via SO_INCOMING_CPU.  From Eric Dumazet.

18) Limit GSO packets to half the current congestion window, from Eric
    Dumazet.

19) Add a generic helper so that all drivers set their RSS keys in a
    consistent way, from Eric Dumazet.

20) Add xmit_more support to enic driver, from Govindarajulu
    Varadarajan.

21) Add VLAN packet scheduler action, from Jiri Pirko.

22) Support configurable RSS hash functions via ethtool, from Eyal
    Perry.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1820 commits)
  Fix race condition between vxlan_sock_add and vxlan_sock_release
  net/macb: fix compilation warning for print_hex_dump() called with skb->mac_header
  net/mlx4: Add support for A0 steering
  net/mlx4: Refactor QUERY_PORT
  net/mlx4_core: Add explicit error message when rule doesn't meet configuration
  net/mlx4: Add A0 hybrid steering
  net/mlx4: Add mlx4_bitmap zone allocator
  net/mlx4: Add a check if there are too many reserved QPs
  net/mlx4: Change QP allocation scheme
  net/mlx4_core: Use tasklet for user-space CQ completion events
  net/mlx4_core: Mask out host side virtualization features for guests
  net/mlx4_en: Set csum level for encapsulated packets
  be2net: Export tunnel offloads only when a VxLAN tunnel is created
  gianfar: Fix dma check map error when DMA_API_DEBUG is enabled
  cxgb4/csiostor: Don't use MASTER_MUST for fw_hello call
  net: fec: only enable mdio interrupt before phy device link up
  net: fec: clear all interrupt events to support i.MX6SX
  net: fec: reset fep link status in suspend function
  net: sock: fix access via invalid file descriptor
  net: introduce helper macro for_each_cmsghdr
  ...

12 files changed, 172 insertions, 148 deletions

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index fb61f6685809..4b8c6116c058 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -472,13 +472,13 @@ static void send_flowc(struct c4iw_ep *ep, struct sk_buff *skb)
 	skb = get_skb(skb, flowclen, GFP_KERNEL);
 	flowc = (struct fw_flowc_wr *)__skb_put(skb, flowclen);
 
-	flowc->op_to_nparams = cpu_to_be32(FW_WR_OP(FW_FLOWC_WR) |
-					   FW_FLOWC_WR_NPARAMS(8));
-	flowc->flowid_len16 = cpu_to_be32(FW_WR_LEN16(DIV_ROUND_UP(flowclen,
-					  16)) | FW_WR_FLOWID(ep->hwtid));
+	flowc->op_to_nparams = cpu_to_be32(FW_WR_OP_V(FW_FLOWC_WR) |
+					   FW_FLOWC_WR_NPARAMS_V(8));
+	flowc->flowid_len16 = cpu_to_be32(FW_WR_LEN16_V(DIV_ROUND_UP(flowclen,
+					  16)) | FW_WR_FLOWID_V(ep->hwtid));
 
 	flowc->mnemval[0].mnemonic = FW_FLOWC_MNEM_PFNVFN;
-	flowc->mnemval[0].val = cpu_to_be32(FW_PFVF_CMD_PFN
+	flowc->mnemval[0].val = cpu_to_be32(FW_PFVF_CMD_PFN_V
 					    (ep->com.dev->rdev.lldi.pf));
 	flowc->mnemval[1].mnemonic = FW_FLOWC_MNEM_CH;
 	flowc->mnemval[1].val = cpu_to_be32(ep->tx_chan);
@@ -649,31 +649,31 @@ static int send_connect(struct c4iw_ep *ep)
 	 * remainder will be specified in the rx_data_ack.
 	 */
 	win = ep->rcv_win >> 10;
-	if (win > RCV_BUFSIZ_MASK)
-		win = RCV_BUFSIZ_MASK;
+	if (win > RCV_BUFSIZ_M)
+		win = RCV_BUFSIZ_M;
 
 	opt0 = (nocong ? NO_CONG(1) : 0) |
-	       KEEP_ALIVE(1) |
+	       KEEP_ALIVE_F |
 	       DELACK(1) |
-	       WND_SCALE(wscale) |
-	       MSS_IDX(mtu_idx) |
-	       L2T_IDX(ep->l2t->idx) |
-	       TX_CHAN(ep->tx_chan) |
-	       SMAC_SEL(ep->smac_idx) |
+	       WND_SCALE_V(wscale) |
+	       MSS_IDX_V(mtu_idx) |
+	       L2T_IDX_V(ep->l2t->idx) |
+	       TX_CHAN_V(ep->tx_chan) |
+	       SMAC_SEL_V(ep->smac_idx) |
 	       DSCP(ep->tos) |
-	       ULP_MODE(ULP_MODE_TCPDDP) |
-	       RCV_BUFSIZ(win);
-	opt2 = RX_CHANNEL(0) |
+	       ULP_MODE_V(ULP_MODE_TCPDDP) |
+	       RCV_BUFSIZ_V(win);
+	opt2 = RX_CHANNEL_V(0) |
 	       CCTRL_ECN(enable_ecn) |
-	       RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);
+	       RSS_QUEUE_VALID_F | RSS_QUEUE_V(ep->rss_qid);
 	if (enable_tcp_timestamps)
 		opt2 |= TSTAMPS_EN(1);
 	if (enable_tcp_sack)
 		opt2 |= SACK_EN(1);
 	if (wscale && enable_tcp_window_scaling)
-		opt2 |= WND_SCALE_EN(1);
+		opt2 |= WND_SCALE_EN_F;
 	if (is_t5(ep->com.dev->rdev.lldi.adapter_type)) {
-		opt2 |= T5_OPT_2_VALID;
+		opt2 |= T5_OPT_2_VALID_F;
 		opt2 |= V_CONG_CNTRL(CONG_ALG_TAHOE);
 		opt2 |= CONG_CNTRL_VALID; /* OPT_2_ISS for T5 */
 	}
@@ -736,7 +736,7 @@ static int send_connect(struct c4iw_ep *ep)
 			t5_req->local_ip = la->sin_addr.s_addr;
 			t5_req->peer_ip = ra->sin_addr.s_addr;
 			t5_req->opt0 = cpu_to_be64(opt0);
-			t5_req->params = cpu_to_be64(V_FILTER_TUPLE(
+			t5_req->params = cpu_to_be64(FILTER_TUPLE_V(
 						     cxgb4_select_ntuple(
 					     ep->com.dev->rdev.lldi.ports[0],
 					     ep->l2t)));
@@ -762,7 +762,7 @@ static int send_connect(struct c4iw_ep *ep)
 			t5_req6->peer_ip_lo = *((__be64 *)
 						(ra6->sin6_addr.s6_addr + 8));
 			t5_req6->opt0 = cpu_to_be64(opt0);
-			t5_req6->params = cpu_to_be64(V_FILTER_TUPLE(
+			t5_req6->params = cpu_to_be64(FILTER_TUPLE_V(
 							cxgb4_select_ntuple(
 						ep->com.dev->rdev.lldi.ports[0],
 						ep->l2t)));
@@ -803,16 +803,16 @@ static void send_mpa_req(struct c4iw_ep *ep, struct sk_buff *skb,
 	req = (struct fw_ofld_tx_data_wr *)skb_put(skb, wrlen);
 	memset(req, 0, wrlen);
 	req->op_to_immdlen = cpu_to_be32(
-		FW_WR_OP(FW_OFLD_TX_DATA_WR) |
-		FW_WR_COMPL(1) |
-		FW_WR_IMMDLEN(mpalen));
+		FW_WR_OP_V(FW_OFLD_TX_DATA_WR) |
+		FW_WR_COMPL_F |
+		FW_WR_IMMDLEN_V(mpalen));
 	req->flowid_len16 = cpu_to_be32(
-		FW_WR_FLOWID(ep->hwtid) |
-		FW_WR_LEN16(wrlen >> 4));
+		FW_WR_FLOWID_V(ep->hwtid) |
+		FW_WR_LEN16_V(wrlen >> 4));
 	req->plen = cpu_to_be32(mpalen);
 	req->tunnel_to_proxy = cpu_to_be32(
-		FW_OFLD_TX_DATA_WR_FLUSH(1) |
-		FW_OFLD_TX_DATA_WR_SHOVE(1));
+		FW_OFLD_TX_DATA_WR_FLUSH_F |
+		FW_OFLD_TX_DATA_WR_SHOVE_F);
 
 	mpa = (struct mpa_message *)(req + 1);
 	memcpy(mpa->key, MPA_KEY_REQ, sizeof(mpa->key));
@@ -897,16 +897,16 @@ static int send_mpa_reject(struct c4iw_ep *ep, const void *pdata, u8 plen)
 	req = (struct fw_ofld_tx_data_wr *)skb_put(skb, wrlen);
 	memset(req, 0, wrlen);
 	req->op_to_immdlen = cpu_to_be32(
-		FW_WR_OP(FW_OFLD_TX_DATA_WR) |
-		FW_WR_COMPL(1) |
-		FW_WR_IMMDLEN(mpalen));
+		FW_WR_OP_V(FW_OFLD_TX_DATA_WR) |
+		FW_WR_COMPL_F |
+		FW_WR_IMMDLEN_V(mpalen));
 	req->flowid_len16 = cpu_to_be32(
-		FW_WR_FLOWID(ep->hwtid) |
-		FW_WR_LEN16(wrlen >> 4));
+		FW_WR_FLOWID_V(ep->hwtid) |
+		FW_WR_LEN16_V(wrlen >> 4));
 	req->plen = cpu_to_be32(mpalen);
 	req->tunnel_to_proxy = cpu_to_be32(
-		FW_OFLD_TX_DATA_WR_FLUSH(1) |
-		FW_OFLD_TX_DATA_WR_SHOVE(1));
+		FW_OFLD_TX_DATA_WR_FLUSH_F |
+		FW_OFLD_TX_DATA_WR_SHOVE_F);
 
 	mpa = (struct mpa_message *)(req + 1);
 	memset(mpa, 0, sizeof(*mpa));
@@ -977,16 +977,16 @@ static int send_mpa_reply(struct c4iw_ep *ep, const void *pdata, u8 plen)
 	req = (struct fw_ofld_tx_data_wr *) skb_put(skb, wrlen);
 	memset(req, 0, wrlen);
 	req->op_to_immdlen = cpu_to_be32(
-		FW_WR_OP(FW_OFLD_TX_DATA_WR) |
-		FW_WR_COMPL(1) |
-		FW_WR_IMMDLEN(mpalen));
+		FW_WR_OP_V(FW_OFLD_TX_DATA_WR) |
+		FW_WR_COMPL_F |
+		FW_WR_IMMDLEN_V(mpalen));
 	req->flowid_len16 = cpu_to_be32(
-		FW_WR_FLOWID(ep->hwtid) |
-		FW_WR_LEN16(wrlen >> 4));
+		FW_WR_FLOWID_V(ep->hwtid) |
+		FW_WR_LEN16_V(wrlen >> 4));
 	req->plen = cpu_to_be32(mpalen);
 	req->tunnel_to_proxy = cpu_to_be32(
-		FW_OFLD_TX_DATA_WR_FLUSH(1) |
-		FW_OFLD_TX_DATA_WR_SHOVE(1));
+		FW_OFLD_TX_DATA_WR_FLUSH_F |
+		FW_OFLD_TX_DATA_WR_SHOVE_F);
 
 	mpa = (struct mpa_message *)(req + 1);
 	memset(mpa, 0, sizeof(*mpa));
@@ -1249,15 +1249,15 @@ static int update_rx_credits(struct c4iw_ep *ep, u32 credits)
 	 * due to the limit in the number of bits in the RCV_BUFSIZ field,
 	 * then add the overage in to the credits returned.
 	 */
-	if (ep->rcv_win > RCV_BUFSIZ_MASK * 1024)
-		credits += ep->rcv_win - RCV_BUFSIZ_MASK * 1024;
+	if (ep->rcv_win > RCV_BUFSIZ_M * 1024)
+		credits += ep->rcv_win - RCV_BUFSIZ_M * 1024;
 
 	req = (struct cpl_rx_data_ack *) skb_put(skb, wrlen);
 	memset(req, 0, wrlen);
 	INIT_TP_WR(req, ep->hwtid);
 	OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_RX_DATA_ACK,
 						    ep->hwtid));
-	req->credit_dack = cpu_to_be32(credits | RX_FORCE_ACK(1) |
+	req->credit_dack = cpu_to_be32(credits | RX_FORCE_ACK_F |
 				       F_RX_DACK_CHANGE |
 				       V_RX_DACK_MODE(dack_mode));
 	set_wr_txq(skb, CPL_PRIORITY_ACK, ep->ctrlq_idx);
@@ -1751,7 +1751,7 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 	req = (struct fw_ofld_connection_wr *)__skb_put(skb, sizeof(*req));
 	memset(req, 0, sizeof(*req));
 	req->op_compl = htonl(V_WR_OP(FW_OFLD_CONNECTION_WR));
-	req->len16_pkd = htonl(FW_WR_LEN16(DIV_ROUND_UP(sizeof(*req), 16)));
+	req->len16_pkd = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*req), 16)));
 	req->le.filter = cpu_to_be32(cxgb4_select_ntuple(
 				     ep->com.dev->rdev.lldi.ports[0],
 				     ep->l2t));
@@ -1762,10 +1762,10 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 	req->le.pport = sin->sin_port;
 	req->le.u.ipv4.pip = sin->sin_addr.s_addr;
 	req->tcb.t_state_to_astid =
-			htonl(V_FW_OFLD_CONNECTION_WR_T_STATE(TCP_SYN_SENT) |
-			V_FW_OFLD_CONNECTION_WR_ASTID(atid));
+			htonl(FW_OFLD_CONNECTION_WR_T_STATE_V(TCP_SYN_SENT) |
+			FW_OFLD_CONNECTION_WR_ASTID_V(atid));
 	req->tcb.cplrxdataack_cplpassacceptrpl =
-			htons(F_FW_OFLD_CONNECTION_WR_CPLRXDATAACK);
+			htons(FW_OFLD_CONNECTION_WR_CPLRXDATAACK_F);
 	req->tcb.tx_max = (__force __be32) jiffies;
 	req->tcb.rcv_adv = htons(1);
 	best_mtu(ep->com.dev->rdev.lldi.mtus, ep->mtu, &mtu_idx,
@@ -1778,34 +1778,34 @@ static void send_fw_act_open_req(struct c4iw_ep *ep, unsigned int atid)
 	 * remainder will be specified in the rx_data_ack.
 	 */
 	win = ep->rcv_win >> 10;
-	if (win > RCV_BUFSIZ_MASK)
-		win = RCV_BUFSIZ_MASK;
+	if (win > RCV_BUFSIZ_M)
+		win = RCV_BUFSIZ_M;
 
 	req->tcb.opt0 = (__force __be64) (TCAM_BYPASS(1) |
 		(nocong ? NO_CONG(1) : 0) |
-		KEEP_ALIVE(1) |
+		KEEP_ALIVE_F |
 		DELACK(1) |
-		WND_SCALE(wscale) |
-		MSS_IDX(mtu_idx) |
-		L2T_IDX(ep->l2t->idx) |
-		TX_CHAN(ep->tx_chan) |
-		SMAC_SEL(ep->smac_idx) |
+		WND_SCALE_V(wscale) |
+		MSS_IDX_V(mtu_idx) |
+		L2T_IDX_V(ep->l2t->idx) |
+		TX_CHAN_V(ep->tx_chan) |
+		SMAC_SEL_V(ep->smac_idx) |
 		DSCP(ep->tos) |
-		ULP_MODE(ULP_MODE_TCPDDP) |
-		RCV_BUFSIZ(win));
+		ULP_MODE_V(ULP_MODE_TCPDDP) |
+		RCV_BUFSIZ_V(win));
 	req->tcb.opt2 = (__force __be32) (PACE(1) |
 		TX_QUEUE(ep->com.dev->rdev.lldi.tx_modq[ep->tx_chan]) |
-		RX_CHANNEL(0) |
+		RX_CHANNEL_V(0) |
 		CCTRL_ECN(enable_ecn) |
-		RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid));
+		RSS_QUEUE_VALID_F | RSS_QUEUE_V(ep->rss_qid));
 	if (enable_tcp_timestamps)
-		req->tcb.opt2 |= (__force __be32) TSTAMPS_EN(1);
+		req->tcb.opt2 |= (__force __be32)TSTAMPS_EN(1);
 	if (enable_tcp_sack)
-		req->tcb.opt2 |= (__force __be32) SACK_EN(1);
+		req->tcb.opt2 |= (__force __be32)SACK_EN(1);
 	if (wscale && enable_tcp_window_scaling)
-		req->tcb.opt2 |= (__force __be32) WND_SCALE_EN(1);
-	req->tcb.opt0 = cpu_to_be64((__force u64) req->tcb.opt0);
-	req->tcb.opt2 = cpu_to_be32((__force u32) req->tcb.opt2);
+		req->tcb.opt2 |= (__force __be32)WND_SCALE_EN_F;
+	req->tcb.opt0 = cpu_to_be64((__force u64)req->tcb.opt0);
+	req->tcb.opt2 = cpu_to_be32((__force u32)req->tcb.opt2);
 	set_wr_txq(skb, CPL_PRIORITY_CONTROL, ep->ctrlq_idx);
 	set_bit(ACT_OFLD_CONN, &ep->com.history);
 	c4iw_l2t_send(&ep->com.dev->rdev, skb, ep->l2t);
@@ -2178,28 +2178,28 @@ static void accept_cr(struct c4iw_ep *ep, struct sk_buff *skb,
 	 * remainder will be specified in the rx_data_ack.
 	 */
 	win = ep->rcv_win >> 10;
-	if (win > RCV_BUFSIZ_MASK)
-		win = RCV_BUFSIZ_MASK;
+	if (win > RCV_BUFSIZ_M)
+		win = RCV_BUFSIZ_M;
 	opt0 = (nocong ? NO_CONG(1) : 0) |
-	       KEEP_ALIVE(1) |
+	       KEEP_ALIVE_F |
 	       DELACK(1) |
-	       WND_SCALE(wscale) |
-	       MSS_IDX(mtu_idx) |
-	       L2T_IDX(ep->l2t->idx) |
-	       TX_CHAN(ep->tx_chan) |
-	       SMAC_SEL(ep->smac_idx) |
+	       WND_SCALE_V(wscale) |
+	       MSS_IDX_V(mtu_idx) |
+	       L2T_IDX_V(ep->l2t->idx) |
+	       TX_CHAN_V(ep->tx_chan) |
+	       SMAC_SEL_V(ep->smac_idx) |
 	       DSCP(ep->tos >> 2) |
-	       ULP_MODE(ULP_MODE_TCPDDP) |
-	       RCV_BUFSIZ(win);
-	opt2 = RX_CHANNEL(0) |
-	       RSS_QUEUE_VALID | RSS_QUEUE(ep->rss_qid);
+	       ULP_MODE_V(ULP_MODE_TCPDDP) |
+	       RCV_BUFSIZ_V(win);
+	opt2 = RX_CHANNEL_V(0) |
+	       RSS_QUEUE_VALID_F | RSS_QUEUE_V(ep->rss_qid);
 
 	if (enable_tcp_timestamps && req->tcpopt.tstamp)
 		opt2 |= TSTAMPS_EN(1);
 	if (enable_tcp_sack && req->tcpopt.sack)
 		opt2 |= SACK_EN(1);
 	if (wscale && enable_tcp_window_scaling)
-		opt2 |= WND_SCALE_EN(1);
+		opt2 |= WND_SCALE_EN_F;
 	if (enable_ecn) {
 		const struct tcphdr *tcph;
 		u32 hlen = ntohl(req->hdr_len);
@@ -2211,7 +2211,7 @@ static void accept_cr(struct c4iw_ep *ep, struct sk_buff *skb,
 	}
 	if (is_t5(ep->com.dev->rdev.lldi.adapter_type)) {
 		u32 isn = (prandom_u32() & ~7UL) - 1;
-		opt2 |= T5_OPT_2_VALID;
+		opt2 |= T5_OPT_2_VALID_F;
 		opt2 |= V_CONG_CNTRL(CONG_ALG_TAHOE);
 		opt2 |= CONG_CNTRL_VALID; /* OPT_2_ISS for T5 */
 		rpl5 = (void *)rpl;
@@ -3537,9 +3537,9 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb,
 	req_skb = alloc_skb(sizeof(struct fw_ofld_connection_wr), GFP_KERNEL);
 	req = (struct fw_ofld_connection_wr *)__skb_put(req_skb, sizeof(*req));
 	memset(req, 0, sizeof(*req));
-	req->op_compl = htonl(V_WR_OP(FW_OFLD_CONNECTION_WR) | FW_WR_COMPL(1));
-	req->len16_pkd = htonl(FW_WR_LEN16(DIV_ROUND_UP(sizeof(*req), 16)));
-	req->le.version_cpl = htonl(F_FW_OFLD_CONNECTION_WR_CPL);
+	req->op_compl = htonl(V_WR_OP(FW_OFLD_CONNECTION_WR) | FW_WR_COMPL_F);
+	req->len16_pkd = htonl(FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*req), 16)));
+	req->le.version_cpl = htonl(FW_OFLD_CONNECTION_WR_CPL_F);
 	req->le.filter = (__force __be32) filter;
 	req->le.lport = lport;
 	req->le.pport = rport;
@@ -3548,16 +3548,16 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb,
 	req->tcb.rcv_nxt = htonl(rcv_isn + 1);
 	req->tcb.rcv_adv = htons(window);
 	req->tcb.t_state_to_astid =
-		 htonl(V_FW_OFLD_CONNECTION_WR_T_STATE(TCP_SYN_RECV) |
-			V_FW_OFLD_CONNECTION_WR_RCV_SCALE(cpl->tcpopt.wsf) |
-			V_FW_OFLD_CONNECTION_WR_ASTID(
+		 htonl(FW_OFLD_CONNECTION_WR_T_STATE_V(TCP_SYN_RECV) |
+			FW_OFLD_CONNECTION_WR_RCV_SCALE_V(cpl->tcpopt.wsf) |
+			FW_OFLD_CONNECTION_WR_ASTID_V(
 			GET_PASS_OPEN_TID(ntohl(cpl->tos_stid))));
 
 	/*
 	 * We store the qid in opt2 which will be used by the firmware
 	 * to send us the wr response.
 	 */
-	req->tcb.opt2 = htonl(V_RSS_QUEUE(rss_qid));
+	req->tcb.opt2 = htonl(RSS_QUEUE_V(rss_qid));
 
 	/*
 	 * We initialize the MSS index in TCB to 0xF.
@@ -3565,7 +3565,7 @@ static void send_fw_pass_open_req(struct c4iw_dev *dev, struct sk_buff *skb,
 	 * TCB picks up the correct value. If this was 0
 	 * TP will ignore any value > 0 for MSS index.
 	 */
-	req->tcb.opt0 = cpu_to_be64(V_MSS_IDX(0xF));
+	req->tcb.opt0 = cpu_to_be64(MSS_IDX_V(0xF));
 	req->cookie = (unsigned long)skb;
 
 	set_wr_txq(req_skb, CPL_PRIORITY_CONTROL, port_id);
diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c
index 0f773e78e080..e9fd3a029296 100644
--- a/drivers/infiniband/hw/cxgb4/cq.c
+++ b/drivers/infiniband/hw/cxgb4/cq.c
@@ -51,9 +51,9 @@ static int destroy_cq(struct c4iw_rdev *rdev, struct t4_cq *cq,
 	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) |
+			FW_WR_OP_V(FW_RI_RES_WR) |
 			V_FW_RI_RES_WR_NRES(1) |
-			FW_WR_COMPL(1));
+			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 = res_wr->res;
@@ -121,9 +121,9 @@ static int create_cq(struct c4iw_rdev *rdev, struct t4_cq *cq,
 	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) |
+			FW_WR_OP_V(FW_RI_RES_WR) |
 			V_FW_RI_RES_WR_NRES(1) |
-			FW_WR_COMPL(1));
+			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 = res_wr->res;
diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c
index ec7a2988a703..0744455cd88b 100644
--- a/drivers/infiniband/hw/cxgb4/mem.c
+++ b/drivers/infiniband/hw/cxgb4/mem.c
@@ -74,18 +74,18 @@ static int _c4iw_write_mem_dma_aligned(struct c4iw_rdev *rdev, u32 addr,
 	req = (struct ulp_mem_io *)__skb_put(skb, wr_len);
 	memset(req, 0, wr_len);
 	INIT_ULPTX_WR(req, wr_len, 0, 0);
-	req->wr.wr_hi = cpu_to_be32(FW_WR_OP(FW_ULPTX_WR) |
-			(wait ? FW_WR_COMPL(1) : 0));
+	req->wr.wr_hi = cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR) |
+			(wait ? FW_WR_COMPL_F : 0));
 	req->wr.wr_lo = wait ? (__force __be64)(unsigned long) &wr_wait : 0L;
-	req->wr.wr_mid = cpu_to_be32(FW_WR_LEN16(DIV_ROUND_UP(wr_len, 16)));
-	req->cmd = cpu_to_be32(ULPTX_CMD(ULP_TX_MEM_WRITE));
+	req->wr.wr_mid = cpu_to_be32(FW_WR_LEN16_V(DIV_ROUND_UP(wr_len, 16)));
+	req->cmd = cpu_to_be32(ULPTX_CMD_V(ULP_TX_MEM_WRITE));
 	req->cmd |= cpu_to_be32(V_T5_ULP_MEMIO_ORDER(1));
-	req->dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN(len>>5));
+	req->dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN_V(len>>5));
 	req->len16 = cpu_to_be32(DIV_ROUND_UP(wr_len-sizeof(req->wr), 16));
-	req->lock_addr = cpu_to_be32(ULP_MEMIO_ADDR(addr));
+	req->lock_addr = cpu_to_be32(ULP_MEMIO_ADDR_V(addr));
 
 	sgl = (struct ulptx_sgl *)(req + 1);
-	sgl->cmd_nsge = cpu_to_be32(ULPTX_CMD(ULP_TX_SC_DSGL) |
+	sgl->cmd_nsge = cpu_to_be32(ULPTX_CMD_V(ULP_TX_SC_DSGL) |
 				    ULPTX_NSGE(1));
 	sgl->len0 = cpu_to_be32(len);
 	sgl->addr0 = cpu_to_be64(data);
@@ -107,12 +107,12 @@ static int _c4iw_write_mem_inline(struct c4iw_rdev *rdev, u32 addr, u32 len,
 	u8 wr_len, *to_dp, *from_dp;
 	int copy_len, num_wqe, i, ret = 0;
 	struct c4iw_wr_wait wr_wait;
-	__be32 cmd = cpu_to_be32(ULPTX_CMD(ULP_TX_MEM_WRITE));
+	__be32 cmd = cpu_to_be32(ULPTX_CMD_V(ULP_TX_MEM_WRITE));
 
 	if (is_t4(rdev->lldi.adapter_type))
-		cmd |= cpu_to_be32(ULP_MEMIO_ORDER(1));
+		cmd |= cpu_to_be32(ULP_MEMIO_ORDER_F);
 	else
-		cmd |= cpu_to_be32(V_T5_ULP_MEMIO_IMM(1));
+		cmd |= cpu_to_be32(T5_ULP_MEMIO_IMM_F);
 
 	addr &= 0x7FFFFFF;
 	PDBG("%s addr 0x%x len %u\n", __func__, addr, len);
@@ -135,23 +135,23 @@ static int _c4iw_write_mem_inline(struct c4iw_rdev *rdev, u32 addr, u32 len,
 		INIT_ULPTX_WR(req, wr_len, 0, 0);
 
 		if (i == (num_wqe-1)) {
-			req->wr.wr_hi = cpu_to_be32(FW_WR_OP(FW_ULPTX_WR) |
-						    FW_WR_COMPL(1));
+			req->wr.wr_hi = cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR) |
+						    FW_WR_COMPL_F);
 			req->wr.wr_lo = (__force __be64)(unsigned long) &wr_wait;
 		} else
-			req->wr.wr_hi = cpu_to_be32(FW_WR_OP(FW_ULPTX_WR));
+			req->wr.wr_hi = cpu_to_be32(FW_WR_OP_V(FW_ULPTX_WR));
 		req->wr.wr_mid = cpu_to_be32(
-				       FW_WR_LEN16(DIV_ROUND_UP(wr_len, 16)));
+				       FW_WR_LEN16_V(DIV_ROUND_UP(wr_len, 16)));
 
 		req->cmd = cmd;
-		req->dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN(
+		req->dlen = cpu_to_be32(ULP_MEMIO_DATA_LEN_V(
 				DIV_ROUND_UP(copy_len, T4_ULPTX_MIN_IO)));
 		req->len16 = cpu_to_be32(DIV_ROUND_UP(wr_len-sizeof(req->wr),
 						      16));
-		req->lock_addr = cpu_to_be32(ULP_MEMIO_ADDR(addr + i * 3));
+		req->lock_addr = cpu_to_be32(ULP_MEMIO_ADDR_V(addr + i * 3));
 
 		sc = (struct ulptx_idata *)(req + 1);
-		sc->cmd_more = cpu_to_be32(ULPTX_CMD(ULP_TX_SC_IMM));
+		sc->cmd_more = cpu_to_be32(ULPTX_CMD_V(ULP_TX_SC_IMM));
 		sc->len = cpu_to_be32(roundup(copy_len, T4_ULPTX_MIN_IO));
 
 		to_dp = (u8 *)(sc + 1);
diff --git a/drivers/infiniband/hw/cxgb4/provider.c b/drivers/infiniband/hw/cxgb4/provider.c
index 72e3b69d1b76..66bd6a2ad83b 100644
--- a/drivers/infiniband/hw/cxgb4/provider.c
+++ b/drivers/infiniband/hw/cxgb4/provider.c
@@ -408,10 +408,10 @@ static ssize_t show_fw_ver(struct device *dev, struct device_attribute *attr,
 	PDBG("%s dev 0x%p\n", __func__, dev);
 
 	return sprintf(buf, "%u.%u.%u.%u\n",
-			FW_HDR_FW_VER_MAJOR_GET(c4iw_dev->rdev.lldi.fw_vers),
-			FW_HDR_FW_VER_MINOR_GET(c4iw_dev->rdev.lldi.fw_vers),
-			FW_HDR_FW_VER_MICRO_GET(c4iw_dev->rdev.lldi.fw_vers),
-			FW_HDR_FW_VER_BUILD_GET(c4iw_dev->rdev.lldi.fw_vers));
+			FW_HDR_FW_VER_MAJOR_G(c4iw_dev->rdev.lldi.fw_vers),
+			FW_HDR_FW_VER_MINOR_G(c4iw_dev->rdev.lldi.fw_vers),
+			FW_HDR_FW_VER_MICRO_G(c4iw_dev->rdev.lldi.fw_vers),
+			FW_HDR_FW_VER_BUILD_G(c4iw_dev->rdev.lldi.fw_vers));
 }
 
 static ssize_t show_hca(struct device *dev, struct device_attribute *attr,
diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index 41cd6882b648..2ed3ece2b2ee 100644
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -271,9 +271,9 @@ static int create_qp(struct c4iw_rdev *rdev, struct t4_wq *wq,
 	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) |
+			FW_WR_OP_V(FW_RI_RES_WR) |
 			V_FW_RI_RES_WR_NRES(2) |
-			FW_WR_COMPL(1));
+			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 = res_wr->res;
@@ -1082,10 +1082,10 @@ static void post_terminate(struct c4iw_qp *qhp, struct t4_cqe *err_cqe,
 
 	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);
@@ -1204,11 +1204,11 @@ static int rdma_fini(struct c4iw_dev *rhp, struct c4iw_qp *qhp,
 	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)));
+		FW_WR_FLOWID_V(ep->hwtid) |
+		FW_WR_LEN16_V(DIV_ROUND_UP(sizeof(*wqe), 16)));
 	wqe->cookie = (unsigned long) &ep->com.wr_wait;
 
 	wqe->u.fini.type = FW_RI_TYPE_FINI;
@@ -1273,11 +1273,11 @@ static int rdma_init(struct c4iw_dev *rhp, struct c4iw_qp *qhp)
 	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;
 
diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 1066eec854a9..a3b70f6c4035 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -233,7 +233,10 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, int entries, int vector
 	if (err)
 		goto err_dbmap;
 
-	cq->mcq.comp  = mlx4_ib_cq_comp;
+	if (context)
+		cq->mcq.tasklet_ctx.comp = mlx4_ib_cq_comp;
+	else
+		cq->mcq.comp = mlx4_ib_cq_comp;
 	cq->mcq.event = mlx4_ib_cq_event;
 
 	if (context)
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 8b72cf392b34..57ecc5b204f3 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -1975,8 +1975,7 @@ static void mlx4_ib_alloc_eqs(struct mlx4_dev *dev, struct mlx4_ib_dev *ibdev)
 	    dev->caps.num_ports > dev->caps.comp_pool)
 		return;
 
-	eq_per_port = rounddown_pow_of_two(dev->caps.comp_pool/
-					dev->caps.num_ports);
+	eq_per_port = dev->caps.comp_pool / dev->caps.num_ports;
 
 	/* Init eq table */
 	added_eqs = 0;
@@ -2228,7 +2227,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 		ibdev->steer_qpn_count = MLX4_IB_UC_MAX_NUM_QPS;
 		err = mlx4_qp_reserve_range(dev, ibdev->steer_qpn_count,
 					    MLX4_IB_UC_STEER_QPN_ALIGN,
-					    &ibdev->steer_qpn_base);
+					    &ibdev->steer_qpn_base, 0);
 		if (err)
 			goto err_counter;
 
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 9c5150c3cb31..cf000b7ad64f 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -802,16 +802,21 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd,
 			}
 		}
 	} else {
-		/* Raw packet QPNs must be aligned to 8 bits. If not, the WQE
-		 * BlueFlame setup flow wrongly causes VLAN insertion. */
+		/* Raw packet QPNs may not have bits 6,7 set in their qp_num;
+		 * otherwise, the WQE BlueFlame setup flow wrongly causes
+		 * VLAN insertion. */
 		if (init_attr->qp_type == IB_QPT_RAW_PACKET)
-			err = mlx4_qp_reserve_range(dev->dev, 1, 1 << 8, &qpn);
+			err = mlx4_qp_reserve_range(dev->dev, 1, 1, &qpn,
+						    (init_attr->cap.max_send_wr ?
+						     MLX4_RESERVE_ETH_BF_QP : 0) |
+						    (init_attr->cap.max_recv_wr ?
+						     MLX4_RESERVE_A0_QP : 0));
 		else
 			if (qp->flags & MLX4_IB_QP_NETIF)
 				err = mlx4_ib_steer_qp_alloc(dev, 1, &qpn);
 			else
 				err = mlx4_qp_reserve_range(dev->dev, 1, 1,
-							    &qpn);
+							    &qpn, 0);
 		if (err)
 			goto err_proxy;
 	}
diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index 10cfce5119a9..c463e7bba5f4 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -805,14 +805,14 @@ struct ib_cq *mlx5_ib_create_cq(struct ib_device *ibdev, int entries,
 		}
 
 
-	mlx5_vfree(cqb);
+	kvfree(cqb);
 	return &cq->ibcq;
 
 err_cmd:
 	mlx5_core_destroy_cq(dev->mdev, &cq->mcq);
 
 err_cqb:
-	mlx5_vfree(cqb);
+	kvfree(cqb);
 	if (context)
 		destroy_cq_user(cq, context);
 	else
@@ -1159,11 +1159,11 @@ int mlx5_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata)
 	}
 	mutex_unlock(&cq->resize_mutex);
 
-	mlx5_vfree(in);
+	kvfree(in);
 	return 0;
 
 ex_alloc:
-	mlx5_vfree(in);
+	kvfree(in);
 
 ex_resize:
 	if (udata)
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 8ee7cb46e059..5a80dd993761 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -159,6 +159,9 @@ static int add_keys(struct mlx5_ib_dev *dev, int c, int num)
 					    sizeof(*in), reg_mr_callback,
 					    mr, &mr->out);
 		if (err) {
+			spin_lock_irq(&ent->lock);
+			ent->pending--;
+			spin_unlock_irq(&ent->lock);
 			mlx5_ib_warn(dev, "create mkey failed %d\n", err);
 			kfree(mr);
 			break;
@@ -853,14 +856,14 @@ static struct mlx5_ib_mr *reg_create(struct ib_pd *pd, u64 virt_addr,
 		goto err_2;
 	}
 	mr->umem = umem;
-	mlx5_vfree(in);
+	kvfree(in);
 
 	mlx5_ib_dbg(dev, "mkey = 0x%x\n", mr->mmr.key);
 
 	return mr;
 
 err_2:
-	mlx5_vfree(in);
+	kvfree(in);
 
 err_1:
 	kfree(mr);
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index e261a53f9a02..1cae1c7132b4 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -647,7 +647,7 @@ err_unmap:
 	mlx5_ib_db_unmap_user(context, &qp->db);
 
 err_free:
-	mlx5_vfree(*in);
+	kvfree(*in);
 
 err_umem:
 	if (qp->umem)
@@ -761,7 +761,7 @@ err_wrid:
 	kfree(qp->rq.wrid);
 
 err_free:
-	mlx5_vfree(*in);
+	kvfree(*in);
 
 err_buf:
 	mlx5_buf_free(dev->mdev, &qp->buf);
@@ -971,7 +971,7 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd,
 		goto err_create;
 	}
 
-	mlx5_vfree(in);
+	kvfree(in);
 	/* Hardware wants QPN written in big-endian order (after
 	 * shifting) for send doorbell.  Precompute this value to save
 	 * a little bit when posting sends.
@@ -988,7 +988,7 @@ err_create:
 	else if (qp->create_type == MLX5_QP_KERNEL)
 		destroy_qp_kernel(dev, qp);
 
-	mlx5_vfree(in);
+	kvfree(in);
 	return err;
 }
 
@@ -1011,9 +1011,14 @@ static void mlx5_ib_lock_cqs(struct mlx5_ib_cq *send_cq, struct mlx5_ib_cq *recv
 			}
 		} else {
 			spin_lock_irq(&send_cq->lock);
+			__acquire(&recv_cq->lock);
 		}
 	} else if (recv_cq) {
 		spin_lock_irq(&recv_cq->lock);
+		__acquire(&send_cq->lock);
+	} else {
+		__acquire(&send_cq->lock);
+		__acquire(&recv_cq->lock);
 	}
 }
 
@@ -1033,10 +1038,15 @@ static void mlx5_ib_unlock_cqs(struct mlx5_ib_cq *send_cq, struct mlx5_ib_cq *re
 				spin_unlock_irq(&recv_cq->lock);
 			}
 		} else {
+			__release(&recv_cq->lock);
 			spin_unlock_irq(&send_cq->lock);
 		}
 	} else if (recv_cq) {
+		__release(&send_cq->lock);
 		spin_unlock_irq(&recv_cq->lock);
+	} else {
+		__release(&recv_cq->lock);
+		__release(&send_cq->lock);
 	}
 }
 
@@ -2411,7 +2421,7 @@ static u8 get_fence(u8 fence, struct ib_send_wr *wr)
 
 static int begin_wqe(struct mlx5_ib_qp *qp, void **seg,
 		     struct mlx5_wqe_ctrl_seg **ctrl,
-		     struct ib_send_wr *wr, int *idx,
+		     struct ib_send_wr *wr, unsigned *idx,
 		     int *size, int nreq)
 {
 	int err = 0;
@@ -2737,6 +2747,8 @@ out:
 
 		if (bf->need_lock)
 			spin_lock(&bf->lock);
+		else
+			__acquire(&bf->lock);
 
 		/* TBD enable WC */
 		if (0 && nreq == 1 && bf->uuarn && inl && size > 1 && size <= bf->buf_size / 16) {
@@ -2753,6 +2765,8 @@ out:
 		bf->offset ^= bf->buf_size;
 		if (bf->need_lock)
 			spin_unlock(&bf->lock);
+		else
+			__release(&bf->lock);
 	}
 
 	spin_unlock_irqrestore(&qp->sq.lock, flags);
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index 97cc1baaa8e3..41fec66217dd 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -141,7 +141,7 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 	return 0;
 
 err_in:
-	mlx5_vfree(*in);
+	kvfree(*in);
 
 err_umem:
 	ib_umem_release(srq->umem);
@@ -209,7 +209,7 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq,
 	return 0;
 
 err_in:
-	mlx5_vfree(*in);
+	kvfree(*in);
 
 err_buf:
 	mlx5_buf_free(dev->mdev, &srq->buf);
@@ -306,7 +306,7 @@ struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd,
 	in->ctx.pd = cpu_to_be32(to_mpd(pd)->pdn);
 	in->ctx.db_record = cpu_to_be64(srq->db.dma);
 	err = mlx5_core_create_srq(dev->mdev, &srq->msrq, in, inlen);
-	mlx5_vfree(in);
+	kvfree(in);
 	if (err) {
 		mlx5_ib_dbg(dev, "create SRQ failed, err %d\n", err);
 		goto err_usr_kern_srq;