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

Pull more infiniband changes from Roland Dreier:
 "Second batch of InfiniBand/RDMA changes for 3.8:
   - cxgb4 changes to fix lookup engine hash collisions
   - mlx4 changes to make flow steering usable
   - fix to IPoIB to avoid pinning dst reference for too long"

* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  RDMA/cxgb4: Fix bug for active and passive LE hash collision path
  RDMA/cxgb4: Fix LE hash collision bug for passive open connection
  RDMA/cxgb4: Fix LE hash collision bug for active open connection
  mlx4_core: Allow choosing flow steering mode
  mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV
  mlx4_core: Fix error flow in the flow steering wrapper
  mlx4_core: Add QPN enforcement for flow steering rules set by VFs
  cxgb4: Add LE hash collision bug fix path in LLD driver
  cxgb4: Add T4 filter support
  IPoIB: Call skb_dst_drop() once skb is enqueued for sending

drivers/infiniband/hw/cxgb4/device.c

@@ -279,6 +279,11 @@ static int stats_show(struct seq_file *seq, void *v)
 	seq_printf(seq, " DB State: %s Transitions %llu\n",
 		   db_state_str[dev->db_state],
 		   dev->rdev.stats.db_state_transitions);
+	seq_printf(seq, "TCAM_FULL: %10llu\n", dev->rdev.stats.tcam_full);
+	seq_printf(seq, "ACT_OFLD_CONN_FAILS: %10llu\n",
+		   dev->rdev.stats.act_ofld_conn_fails);
+	seq_printf(seq, "PAS_OFLD_CONN_FAILS: %10llu\n",
+		   dev->rdev.stats.pas_ofld_conn_fails);
 	return 0;
 }
 
@@ -309,6 +314,9 @@ static ssize_t stats_clear(struct file *file, const char __user *buf,
 	dev->rdev.stats.db_empty = 0;
 	dev->rdev.stats.db_drop = 0;
 	dev->rdev.stats.db_state_transitions = 0;
+	dev->rdev.stats.tcam_full = 0;
+	dev->rdev.stats.act_ofld_conn_fails = 0;
+	dev->rdev.stats.pas_ofld_conn_fails = 0;
 	mutex_unlock(&dev->rdev.stats.lock);
 	return count;
 }
@@ -322,6 +330,113 @@ static const struct file_operations stats_debugfs_fops = {
 	.write   = stats_clear,
 };
 
+static int dump_ep(int id, void *p, void *data)
+{
+	struct c4iw_ep *ep = p;
+	struct c4iw_debugfs_data *epd = data;
+	int space;
+	int cc;
+
+	space = epd->bufsize - epd->pos - 1;
+	if (space == 0)
+		return 1;
+
+	cc = snprintf(epd->buf + epd->pos, space,
+			"ep %p cm_id %p qp %p state %d flags 0x%lx history 0x%lx "
+			"hwtid %d atid %d %pI4:%d <-> %pI4:%d\n",
+			ep, ep->com.cm_id, ep->com.qp, (int)ep->com.state,
+			ep->com.flags, ep->com.history, ep->hwtid, ep->atid,
+			&ep->com.local_addr.sin_addr.s_addr,
+			ntohs(ep->com.local_addr.sin_port),
+			&ep->com.remote_addr.sin_addr.s_addr,
+			ntohs(ep->com.remote_addr.sin_port));
+	if (cc < space)
+		epd->pos += cc;
+	return 0;
+}
+
+static int dump_listen_ep(int id, void *p, void *data)
+{
+	struct c4iw_listen_ep *ep = p;
+	struct c4iw_debugfs_data *epd = data;
+	int space;
+	int cc;
+
+	space = epd->bufsize - epd->pos - 1;
+	if (space == 0)
+		return 1;
+
+	cc = snprintf(epd->buf + epd->pos, space,
+			"ep %p cm_id %p state %d flags 0x%lx stid %d backlog %d "
+			"%pI4:%d\n", ep, ep->com.cm_id, (int)ep->com.state,
+			ep->com.flags, ep->stid, ep->backlog,
+			&ep->com.local_addr.sin_addr.s_addr,
+			ntohs(ep->com.local_addr.sin_port));
+	if (cc < space)
+		epd->pos += cc;
+	return 0;
+}
+
+static int ep_release(struct inode *inode, struct file *file)
+{
+	struct c4iw_debugfs_data *epd = file->private_data;
+	if (!epd) {
+		pr_info("%s null qpd?\n", __func__);
+		return 0;
+	}
+	vfree(epd->buf);
+	kfree(epd);
+	return 0;
+}
+
+static int ep_open(struct inode *inode, struct file *file)
+{
+	struct c4iw_debugfs_data *epd;
+	int ret = 0;
+	int count = 1;
+
+	epd = kmalloc(sizeof(*epd), GFP_KERNEL);
+	if (!epd) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	epd->devp = inode->i_private;
+	epd->pos = 0;
+
+	spin_lock_irq(&epd->devp->lock);
+	idr_for_each(&epd->devp->hwtid_idr, count_idrs, &count);
+	idr_for_each(&epd->devp->atid_idr, count_idrs, &count);
+	idr_for_each(&epd->devp->stid_idr, count_idrs, &count);
+	spin_unlock_irq(&epd->devp->lock);
+
+	epd->bufsize = count * 160;
+	epd->buf = vmalloc(epd->bufsize);
+	if (!epd->buf) {
+		ret = -ENOMEM;
+		goto err1;
+	}
+
+	spin_lock_irq(&epd->devp->lock);
+	idr_for_each(&epd->devp->hwtid_idr, dump_ep, epd);
+	idr_for_each(&epd->devp->atid_idr, dump_ep, epd);
+	idr_for_each(&epd->devp->stid_idr, dump_listen_ep, epd);
+	spin_unlock_irq(&epd->devp->lock);
+
+	file->private_data = epd;
+	goto out;
+err1:
+	kfree(epd);
+out:
+	return ret;
+}
+
+static const struct file_operations ep_debugfs_fops = {
+	.owner   = THIS_MODULE,
+	.open    = ep_open,
+	.release = ep_release,
+	.read    = debugfs_read,
+};
+
 static int setup_debugfs(struct c4iw_dev *devp)
 {
 	struct dentry *de;
@@ -344,6 +459,11 @@ static int setup_debugfs(struct c4iw_dev *devp)
 	if (de && de->d_inode)
 		de->d_inode->i_size = 4096;
 
+	de = debugfs_create_file("eps", S_IWUSR, devp->debugfs_root,
+			(void *)devp, &ep_debugfs_fops);
+	if (de && de->d_inode)
+		de->d_inode->i_size = 4096;
+
 	return 0;
 }
 
@@ -475,6 +595,9 @@ static void c4iw_dealloc(struct uld_ctx *ctx)
 	idr_destroy(&ctx->dev->cqidr);
 	idr_destroy(&ctx->dev->qpidr);
 	idr_destroy(&ctx->dev->mmidr);
+	idr_destroy(&ctx->dev->hwtid_idr);
+	idr_destroy(&ctx->dev->stid_idr);
+	idr_destroy(&ctx->dev->atid_idr);
 	iounmap(ctx->dev->rdev.oc_mw_kva);
 	ib_dealloc_device(&ctx->dev->ibdev);
 	ctx->dev = NULL;
@@ -532,6 +655,9 @@ static struct c4iw_dev *c4iw_alloc(const struct cxgb4_lld_info *infop)
 	idr_init(&devp->cqidr);
 	idr_init(&devp->qpidr);
 	idr_init(&devp->mmidr);
+	idr_init(&devp->hwtid_idr);
+	idr_init(&devp->stid_idr);
+	idr_init(&devp->atid_idr);
 	spin_lock_init(&devp->lock);
 	mutex_init(&devp->rdev.stats.lock);
 	mutex_init(&devp->db_mutex);
@@ -577,14 +703,76 @@ static void *c4iw_uld_add(const struct cxgb4_lld_info *infop)
 	return ctx;
 }
 
+static inline struct sk_buff *copy_gl_to_skb_pkt(const struct pkt_gl *gl,
+						 const __be64 *rsp,
+						 u32 pktshift)
+{
+	struct sk_buff *skb;
+
+	/*
+	 * Allocate space for cpl_pass_accept_req which will be synthesized by
+	 * driver. Once the driver synthesizes the request the skb will go
+	 * through the regular cpl_pass_accept_req processing.
+	 * The math here assumes sizeof cpl_pass_accept_req >= sizeof
+	 * cpl_rx_pkt.
+	 */
+	skb = alloc_skb(gl->tot_len + sizeof(struct cpl_pass_accept_req) +
+			sizeof(struct rss_header) - pktshift, GFP_ATOMIC);
+	if (unlikely(!skb))
+		return NULL;
+
+	 __skb_put(skb, gl->tot_len + sizeof(struct cpl_pass_accept_req) +
+		   sizeof(struct rss_header) - pktshift);
+
+	/*
+	 * This skb will contain:
+	 *   rss_header from the rspq descriptor (1 flit)
+	 *   cpl_rx_pkt struct from the rspq descriptor (2 flits)
+	 *   space for the difference between the size of an
+	 *      rx_pkt and pass_accept_req cpl (1 flit)
+	 *   the packet data from the gl
+	 */
+	skb_copy_to_linear_data(skb, rsp, sizeof(struct cpl_pass_accept_req) +
+				sizeof(struct rss_header));
+	skb_copy_to_linear_data_offset(skb, sizeof(struct rss_header) +
+				       sizeof(struct cpl_pass_accept_req),
+				       gl->va + pktshift,
+				       gl->tot_len - pktshift);
+	return skb;
+}
+
+static inline int recv_rx_pkt(struct c4iw_dev *dev, const struct pkt_gl *gl,
+			   const __be64 *rsp)
+{
+	unsigned int opcode = *(u8 *)rsp;
+	struct sk_buff *skb;
+
+	if (opcode != CPL_RX_PKT)
+		goto out;
+
+	skb = copy_gl_to_skb_pkt(gl , rsp, dev->rdev.lldi.sge_pktshift);
+	if (skb == NULL)
+		goto out;
+
+	if (c4iw_handlers[opcode] == NULL) {
+		pr_info("%s no handler opcode 0x%x...\n", __func__,
+		       opcode);
+		kfree_skb(skb);
+		goto out;
+	}
+	c4iw_handlers[opcode](dev, skb);
+	return 1;
+out:
+	return 0;
+}
+
 static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp,
 			const struct pkt_gl *gl)
 {
 	struct uld_ctx *ctx = handle;
 	struct c4iw_dev *dev = ctx->dev;
 	struct sk_buff *skb;
-	const struct cpl_act_establish *rpl;
-	unsigned int opcode;
+	u8 opcode;
 
 	if (gl == NULL) {
 		/* omit RSS and rsp_ctrl at end of descriptor */
@@ -600,6 +788,18 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp,
 
 		u32 qid = be32_to_cpu(rc->pldbuflen_qid);
 		c4iw_ev_handler(dev, qid);
+		return 0;
+	} else if (unlikely(*(u8 *)rsp != *(u8 *)gl->va)) {
+		if (recv_rx_pkt(dev, gl, rsp))
+			return 0;
+
+		pr_info("%s: unexpected FL contents at %p, " \
+		       "RSS %#llx, FL %#llx, len %u\n",
+		       pci_name(ctx->lldi.pdev), gl->va,
+		       (unsigned long long)be64_to_cpu(*rsp),
+		       (unsigned long long)be64_to_cpu(*(u64 *)gl->va),
+		       gl->tot_len);
+
 		return 0;
 	} else {
 		skb = cxgb4_pktgl_to_skb(gl, 128, 128);
@@ -607,13 +807,11 @@ static int c4iw_uld_rx_handler(void *handle, const __be64 *rsp,
 			goto nomem;
 	}
 
-	rpl = cplhdr(skb);
-	opcode = rpl->ot.opcode;
-
+	opcode = *(u8 *)rsp;
 	if (c4iw_handlers[opcode])
 		c4iw_handlers[opcode](dev, skb);
 	else
-		printk(KERN_INFO "%s no handler opcode 0x%x...\n", __func__,
+		pr_info("%s no handler opcode 0x%x...\n", __func__,
 		       opcode);
 
 	return 0;

drivers/infiniband/hw/cxgb4/iw_cxgb4.h

@@ -130,6 +130,9 @@ struct c4iw_stats {
 	u64  db_empty;
 	u64  db_drop;
 	u64  db_state_transitions;
+	u64  tcam_full;
+	u64  act_ofld_conn_fails;
+	u64  pas_ofld_conn_fails;
 };
 
 struct c4iw_rdev {
@@ -223,6 +226,9 @@ struct c4iw_dev {
 	struct dentry *debugfs_root;
 	enum db_state db_state;
 	int qpcnt;
+	struct idr hwtid_idr;
+	struct idr atid_idr;
+	struct idr stid_idr;
 };
 
 static inline struct c4iw_dev *to_c4iw_dev(struct ib_device *ibdev)
@@ -712,6 +718,31 @@ enum c4iw_ep_flags {
 	CLOSE_SENT		= 3,
 };
 
+enum c4iw_ep_history {
+	ACT_OPEN_REQ            = 0,
+	ACT_OFLD_CONN           = 1,
+	ACT_OPEN_RPL            = 2,
+	ACT_ESTAB               = 3,
+	PASS_ACCEPT_REQ         = 4,
+	PASS_ESTAB              = 5,
+	ABORT_UPCALL            = 6,
+	ESTAB_UPCALL            = 7,
+	CLOSE_UPCALL            = 8,
+	ULP_ACCEPT              = 9,
+	ULP_REJECT              = 10,
+	TIMEDOUT                = 11,
+	PEER_ABORT              = 12,
+	PEER_CLOSE              = 13,
+	CONNREQ_UPCALL          = 14,
+	ABORT_CONN              = 15,
+	DISCONN_UPCALL          = 16,
+	EP_DISC_CLOSE           = 17,
+	EP_DISC_ABORT           = 18,
+	CONN_RPL_UPCALL         = 19,
+	ACT_RETRY_NOMEM         = 20,
+	ACT_RETRY_INUSE         = 21
+};
+
 struct c4iw_ep_common {
 	struct iw_cm_id *cm_id;
 	struct c4iw_qp *qp;
@@ -723,6 +754,7 @@ struct c4iw_ep_common {
 	struct sockaddr_in remote_addr;
 	struct c4iw_wr_wait wr_wait;
 	unsigned long flags;
+	unsigned long history;
 };
 
 struct c4iw_listen_ep {
@@ -760,6 +792,7 @@ struct c4iw_ep {
 	u8 tos;
 	u8 retry_with_mpa_v1;
 	u8 tried_with_mpa_v1;
+	unsigned int retry_count;
 };
 
 static inline struct c4iw_ep *to_ep(struct iw_cm_id *cm_id)

drivers/infiniband/ulp/ipoib/ipoib_cm.c

@@ -752,6 +752,9 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_
 		dev->trans_start = jiffies;
 		++tx->tx_head;
 
+		skb_orphan(skb);
+		skb_dst_drop(skb);
+
 		if (++priv->tx_outstanding == ipoib_sendq_size) {
 			ipoib_dbg(priv, "TX ring 0x%x full, stopping kernel net queue\n",
 				  tx->qp->qp_num);

drivers/infiniband/ulp/ipoib/ipoib_ib.c

@@ -615,8 +615,9 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb,
 
 		address->last_send = priv->tx_head;
 		++priv->tx_head;
-		skb_orphan(skb);
 
+		skb_orphan(skb);
+		skb_dst_drop(skb);
 	}
 
 	if (unlikely(priv->tx_outstanding > MAX_SEND_CQE))

drivers/net/ethernet/chelsio/cxgb4/cxgb4.h

@@ -35,6 +35,8 @@
 #ifndef __CXGB4_H__
 #define __CXGB4_H__
 
+#include "t4_hw.h"
+
 #include <linux/bitops.h>
 #include <linux/cache.h>
 #include <linux/interrupt.h>
@@ -212,6 +214,8 @@ struct tp_err_stats {
 struct tp_params {
 	unsigned int ntxchan;        /* # of Tx channels */
 	unsigned int tre;            /* log2 of core clocks per TP tick */
+	unsigned short tx_modq_map;  /* TX modulation scheduler queue to */
+				     /* channel map */
 
 	uint32_t dack_re;            /* DACK timer resolution */
 	unsigned short tx_modq[NCHAN];	/* channel to modulation queue map */
@@ -526,6 +530,7 @@ struct adapter {
 	struct net_device *port[MAX_NPORTS];
 	u8 chan_map[NCHAN];                   /* channel -> port map */
 
+	u32 filter_mode;
 	unsigned int l2t_start;
 	unsigned int l2t_end;
 	struct l2t_data *l2t;
@@ -545,6 +550,129 @@ struct adapter {
 	spinlock_t stats_lock;
 };
 
+/* Defined bit width of user definable filter tuples
+ */
+#define ETHTYPE_BITWIDTH 16
+#define FRAG_BITWIDTH 1
+#define MACIDX_BITWIDTH 9
+#define FCOE_BITWIDTH 1
+#define IPORT_BITWIDTH 3
+#define MATCHTYPE_BITWIDTH 3
+#define PROTO_BITWIDTH 8
+#define TOS_BITWIDTH 8
+#define PF_BITWIDTH 8
+#define VF_BITWIDTH 8
+#define IVLAN_BITWIDTH 16
+#define OVLAN_BITWIDTH 16
+
+/* Filter matching rules.  These consist of a set of ingress packet field
+ * (value, mask) tuples.  The associated ingress packet field matches the
+ * tuple when ((field & mask) == value).  (Thus a wildcard "don't care" field
+ * rule can be constructed by specifying a tuple of (0, 0).)  A filter rule
+ * matches an ingress packet when all of the individual individual field
+ * matching rules are true.
+ *
+ * Partial field masks are always valid, however, while it may be easy to
+ * understand their meanings for some fields (e.g. IP address to match a
+ * subnet), for others making sensible partial masks is less intuitive (e.g.
+ * MPS match type) ...
+ *
+ * Most of the following data structures are modeled on T4 capabilities.
+ * Drivers for earlier chips use the subsets which make sense for those chips.
+ * We really need to come up with a hardware-independent mechanism to
+ * represent hardware filter capabilities ...
+ */
+struct ch_filter_tuple {
+	/* Compressed header matching field rules.  The TP_VLAN_PRI_MAP
+	 * register selects which of these fields will participate in the
+	 * filter match rules -- up to a maximum of 36 bits.  Because
+	 * TP_VLAN_PRI_MAP is a global register, all filters must use the same
+	 * set of fields.
+	 */
+	uint32_t ethtype:ETHTYPE_BITWIDTH;      /* Ethernet type */
+	uint32_t frag:FRAG_BITWIDTH;            /* IP fragmentation header */
+	uint32_t ivlan_vld:1;                   /* inner VLAN valid */
+	uint32_t ovlan_vld:1;                   /* outer VLAN valid */
+	uint32_t pfvf_vld:1;                    /* PF/VF valid */
+	uint32_t macidx:MACIDX_BITWIDTH;        /* exact match MAC index */
+	uint32_t fcoe:FCOE_BITWIDTH;            /* FCoE packet */
+	uint32_t iport:IPORT_BITWIDTH;          /* ingress port */
+	uint32_t matchtype:MATCHTYPE_BITWIDTH;  /* MPS match type */
+	uint32_t proto:PROTO_BITWIDTH;          /* protocol type */
+	uint32_t tos:TOS_BITWIDTH;              /* TOS/Traffic Type */
+	uint32_t pf:PF_BITWIDTH;                /* PCI-E PF ID */
+	uint32_t vf:VF_BITWIDTH;                /* PCI-E VF ID */
+	uint32_t ivlan:IVLAN_BITWIDTH;          /* inner VLAN */
+	uint32_t ovlan:OVLAN_BITWIDTH;          /* outer VLAN */
+
+	/* Uncompressed header matching field rules.  These are always
+	 * available for field rules.
+	 */
+	uint8_t lip[16];        /* local IP address (IPv4 in [3:0]) */
+	uint8_t fip[16];        /* foreign IP address (IPv4 in [3:0]) */
+	uint16_t lport;         /* local port */
+	uint16_t fport;         /* foreign port */
+};
+
+/* A filter ioctl command.
+ */
+struct ch_filter_specification {
+	/* Administrative fields for filter.
+	 */
+	uint32_t hitcnts:1;     /* count filter hits in TCB */
+	uint32_t prio:1;        /* filter has priority over active/server */
+
+	/* Fundamental filter typing.  This is the one element of filter
+	 * matching that doesn't exist as a (value, mask) tuple.
+	 */
+	uint32_t type:1;        /* 0 => IPv4, 1 => IPv6 */
+
+	/* Packet dispatch information.  Ingress packets which match the
+	 * filter rules will be dropped, passed to the host or switched back
+	 * out as egress packets.
+	 */
+	uint32_t action:2;      /* drop, pass, switch */
+
+	uint32_t rpttid:1;      /* report TID in RSS hash field */
+
+	uint32_t dirsteer:1;    /* 0 => RSS, 1 => steer to iq */
+	uint32_t iq:10;         /* ingress queue */
+
+	uint32_t maskhash:1;    /* dirsteer=0: store RSS hash in TCB */
+	uint32_t dirsteerhash:1;/* dirsteer=1: 0 => TCB contains RSS hash */
+				/*             1 => TCB contains IQ ID */
+
+	/* Switch proxy/rewrite fields.  An ingress packet which matches a
+	 * filter with "switch" set will be looped back out as an egress
+	 * packet -- potentially with some Ethernet header rewriting.
+	 */
+	uint32_t eport:2;       /* egress port to switch packet out */
+	uint32_t newdmac:1;     /* rewrite destination MAC address */
+	uint32_t newsmac:1;     /* rewrite source MAC address */
+	uint32_t newvlan:2;     /* rewrite VLAN Tag */
+	uint8_t dmac[ETH_ALEN]; /* new destination MAC address */
+	uint8_t smac[ETH_ALEN]; /* new source MAC address */
+	uint16_t vlan;          /* VLAN Tag to insert */
+
+	/* Filter rule value/mask pairs.
+	 */
+	struct ch_filter_tuple val;
+	struct ch_filter_tuple mask;
+};
+
+enum {
+	FILTER_PASS = 0,        /* default */
+	FILTER_DROP,
+	FILTER_SWITCH
+};
+
+enum {
+	VLAN_NOCHANGE = 0,      /* default */
+	VLAN_REMOVE,
+	VLAN_INSERT,
+	VLAN_REWRITE
+};
+
 static inline u32 t4_read_reg(struct adapter *adap, u32 reg_addr)
 {
 	return readl(adap->regs + reg_addr);
@@ -701,6 +829,12 @@ static inline int t4_wr_mbox_ns(struct adapter *adap, int mbox, const void *cmd,
 void t4_write_indirect(struct adapter *adap, unsigned int addr_reg,
 		       unsigned int data_reg, const u32 *vals,
 		       unsigned int nregs, unsigned int start_idx);
+void t4_read_indirect(struct adapter *adap, unsigned int addr_reg,
+		      unsigned int data_reg, u32 *vals, unsigned int nregs,
+		      unsigned int start_idx);
+
+struct fw_filter_wr;
+
 void t4_intr_enable(struct adapter *adapter);
 void t4_intr_disable(struct adapter *adapter);
 int t4_slow_intr_handler(struct adapter *adapter);
@@ -737,6 +871,8 @@ void t4_tp_get_tcp_stats(struct adapter *adap, struct tp_tcp_stats *v4,
 void t4_load_mtus(struct adapter *adap, const unsigned short *mtus,
 		  const unsigned short *alpha, const unsigned short *beta);
 
+void t4_mk_filtdelwr(unsigned int ftid, struct fw_filter_wr *wr, int qid);
+
 void t4_wol_magic_enable(struct adapter *adap, unsigned int port,
 			 const u8 *addr);
 int t4_wol_pat_enable(struct adapter *adap, unsigned int port, unsigned int map,

drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h

@@ -38,6 +38,7 @@
 #include <linux/cache.h>
 #include <linux/spinlock.h>
 #include <linux/skbuff.h>
+#include <linux/inetdevice.h>
 #include <linux/atomic.h>
 
 /* CPL message priority levels */
@@ -97,7 +98,9 @@ struct tid_info {
 
 	union aopen_entry *atid_tab;
 	unsigned int natids;
+	unsigned int atid_base;
 
+	struct filter_entry *ftid_tab;
 	unsigned int nftids;
 	unsigned int ftid_base;
 	unsigned int aftid_base;
@@ -129,7 +132,7 @@ static inline void *lookup_atid(const struct tid_info *t, unsigned int atid)
 static inline void *lookup_stid(const struct tid_info *t, unsigned int stid)
 {
 	stid -= t->stid_base;
-	return stid < t->nstids ? t->stid_tab[stid].data : NULL;
+	return stid < (t->nstids + t->nsftids) ? t->stid_tab[stid].data : NULL;
 }
 
 static inline void cxgb4_insert_tid(struct tid_info *t, void *data,
@@ -141,6 +144,7 @@ static inline void cxgb4_insert_tid(struct tid_info *t, void *data,
 
 int cxgb4_alloc_atid(struct tid_info *t, void *data);
 int cxgb4_alloc_stid(struct tid_info *t, int family, void *data);
+int cxgb4_alloc_sftid(struct tid_info *t, int family, void *data);
 void cxgb4_free_atid(struct tid_info *t, unsigned int atid);
 void cxgb4_free_stid(struct tid_info *t, unsigned int stid, int family);
 void cxgb4_remove_tid(struct tid_info *t, unsigned int qid, unsigned int tid);
@@ -148,8 +152,14 @@ void cxgb4_remove_tid(struct tid_info *t, unsigned int qid, unsigned int tid);
 struct in6_addr;
 
 int cxgb4_create_server(const struct net_device *dev, unsigned int stid,
-			__be32 sip, __be16 sport, unsigned int queue);
-
+			__be32 sip, __be16 sport, __be16 vlan,
+			unsigned int queue);
+int cxgb4_create_server_filter(const struct net_device *dev, unsigned int stid,
+			       __be32 sip, __be16 sport, __be16 vlan,
+			       unsigned int queue,
+			       unsigned char port, unsigned char mask);
+int cxgb4_remove_server_filter(const struct net_device *dev, unsigned int stid,
+			       unsigned int queue, bool ipv6);
 static inline void set_wr_txq(struct sk_buff *skb, int prio, int queue)
 {
 	skb_set_queue_mapping(skb, (queue << 1) | prio);
@@ -221,9 +231,16 @@ struct cxgb4_lld_info {
 	unsigned int iscsi_iolen;            /* iSCSI max I/O length */
 	unsigned short udb_density;          /* # of user DB/page */
 	unsigned short ucq_density;          /* # of user CQs/page */
+	unsigned short filt_mode;            /* filter optional components */
+	unsigned short tx_modq[NCHAN];       /* maps each tx channel to a */
+					     /* scheduler queue */
 	void __iomem *gts_reg;               /* address of GTS register */
 	void __iomem *db_reg;                /* address of kernel doorbell */
 	int dbfifo_int_thresh;		     /* doorbell fifo int threshold */
+	unsigned int sge_pktshift;           /* Padding between CPL and */
+					     /*	packet data */
+	bool enable_fw_ofld_conn;            /* Enable connection through fw */
+					     /* WR */
 };
 
 struct cxgb4_uld_info {

drivers/net/ethernet/chelsio/cxgb4/l2t.c

@@ -484,6 +484,38 @@ void t4_l2t_update(struct adapter *adap, struct neighbour *neigh)
 		handle_failed_resolution(adap, arpq);
 }
 
+/* Allocate an L2T entry for use by a switching rule.  Such need to be
+ * explicitly freed and while busy they are not on any hash chain, so normal
+ * address resolution updates do not see them.
+ */
+struct l2t_entry *t4_l2t_alloc_switching(struct l2t_data *d)
+{
+	struct l2t_entry *e;
+
+	write_lock_bh(&d->lock);
+	e = alloc_l2e(d);
+	if (e) {
+		spin_lock(&e->lock);          /* avoid race with t4_l2t_free */
+		e->state = L2T_STATE_SWITCHING;
+		atomic_set(&e->refcnt, 1);
+		spin_unlock(&e->lock);
+	}
+	write_unlock_bh(&d->lock);
+	return e;
+}
+
+/* Sets/updates the contents of a switching L2T entry that has been allocated
+ * with an earlier call to @t4_l2t_alloc_switching.
+ */
+int t4_l2t_set_switching(struct adapter *adap, struct l2t_entry *e, u16 vlan,
+		u8 port, u8 *eth_addr)
+{
+	e->vlan = vlan;
+	e->lport = port;
+	memcpy(e->dmac, eth_addr, ETH_ALEN);
+	return write_l2e(adap, e, 0);
+}
+
 struct l2t_data *t4_init_l2t(void)
 {
 	int i;

drivers/net/ethernet/chelsio/cxgb4/l2t.h

@@ -100,6 +100,9 @@ struct l2t_entry *cxgb4_l2t_get(struct l2t_data *d, struct neighbour *neigh,
 				unsigned int priority);
 
 void t4_l2t_update(struct adapter *adap, struct neighbour *neigh);
+struct l2t_entry *t4_l2t_alloc_switching(struct l2t_data *d);
+int t4_l2t_set_switching(struct adapter *adap, struct l2t_entry *e, u16 vlan,
+			 u8 port, u8 *eth_addr);
 struct l2t_data *t4_init_l2t(void);
 void do_l2t_write_rpl(struct adapter *p, const struct cpl_l2t_write_rpl *rpl);
 

drivers/net/ethernet/chelsio/cxgb4/t4_hw.c

@@ -109,7 +109,7 @@ void t4_set_reg_field(struct adapter *adapter, unsigned int addr, u32 mask,
  *	Reads registers that are accessed indirectly through an address/data
  *	register pair.
  */
-static void t4_read_indirect(struct adapter *adap, unsigned int addr_reg,
+void t4_read_indirect(struct adapter *adap, unsigned int addr_reg,
 			     unsigned int data_reg, u32 *vals,
 			     unsigned int nregs, unsigned int start_idx)
 {
@@ -2268,6 +2268,26 @@ int t4_wol_pat_enable(struct adapter *adap, unsigned int port, unsigned int map,
 	return 0;
 }
 
+/*     t4_mk_filtdelwr - create a delete filter WR
+ *     @ftid: the filter ID
+ *     @wr: the filter work request to populate
+ *     @qid: ingress queue to receive the delete notification
+ *
+ *     Creates a filter work request to delete the supplied filter.  If @qid is
+ *     negative the delete notification is suppressed.
+ */
+void t4_mk_filtdelwr(unsigned int ftid, struct fw_filter_wr *wr, int qid)
+{
+	memset(wr, 0, sizeof(*wr));
+	wr->op_pkd = htonl(FW_WR_OP(FW_FILTER_WR));
+	wr->len16_pkd = htonl(FW_WR_LEN16(sizeof(*wr) / 16));
+	wr->tid_to_iq = htonl(V_FW_FILTER_WR_TID(ftid) |
+			V_FW_FILTER_WR_NOREPLY(qid < 0));
+	wr->del_filter_to_l2tix = htonl(F_FW_FILTER_WR_DEL_FILTER);
+	if (qid >= 0)
+		wr->rx_chan_rx_rpl_iq = htons(V_FW_FILTER_WR_RX_RPL_IQ(qid));
+}
+
 #define INIT_CMD(var, cmd, rd_wr) do { \
 	(var).op_to_write = htonl(FW_CMD_OP(FW_##cmd##_CMD) | \
 				  FW_CMD_REQUEST | FW_CMD_##rd_wr); \

drivers/net/ethernet/chelsio/cxgb4/t4_msg.h

@@ -193,8 +193,24 @@ struct work_request_hdr {
 	__be64 wr_lo;
 };
 
+/* wr_hi fields */
+#define S_WR_OP    24
+#define V_WR_OP(x) ((__u64)(x) << S_WR_OP)
+
 #define WR_HDR struct work_request_hdr wr
 
+/* option 0 fields */
+#define S_MSS_IDX    60
+#define M_MSS_IDX    0xF
+#define V_MSS_IDX(x) ((__u64)(x) << S_MSS_IDX)
+#define G_MSS_IDX(x) (((x) >> S_MSS_IDX) & M_MSS_IDX)
+
+/* option 2 fields */
+#define S_RSS_QUEUE    0
+#define M_RSS_QUEUE    0x3FF
+#define V_RSS_QUEUE(x) ((x) << S_RSS_QUEUE)
+#define G_RSS_QUEUE(x) (((x) >> S_RSS_QUEUE) & M_RSS_QUEUE)
+
 struct cpl_pass_open_req {
 	WR_HDR;
 	union opcode_tid ot;
@@ -204,12 +220,14 @@ struct cpl_pass_open_req {
 	__be32 peer_ip;
 	__be64 opt0;
 #define TX_CHAN(x)    ((x) << 2)
+#define NO_CONG(x)    ((x) << 4)
 #define DELACK(x)     ((x) << 5)
 #define ULP_MODE(x)   ((x) << 8)
 #define RCV_BUFSIZ(x) ((x) << 12)
 #define DSCP(x)       ((x) << 22)
 #define SMAC_SEL(x)   ((u64)(x) << 28)
 #define L2T_IDX(x)    ((u64)(x) << 36)
+#define TCAM_BYPASS(x) ((u64)(x) << 48)
 #define NAGLE(x)      ((u64)(x) << 49)
 #define WND_SCALE(x)  ((u64)(x) << 50)
 #define KEEP_ALIVE(x) ((u64)(x) << 54)
@@ -247,8 +265,10 @@ struct cpl_pass_accept_rpl {
 #define RSS_QUEUE_VALID      (1 << 10)
 #define RX_COALESCE_VALID(x) ((x) << 11)
 #define RX_COALESCE(x)       ((x) << 12)
+#define PACE(x)	      ((x) << 16)
 #define TX_QUEUE(x)          ((x) << 23)
 #define RX_CHANNEL(x)        ((x) << 26)
+#define CCTRL_ECN(x)         ((x) << 27)
 #define WND_SCALE_EN(x)      ((x) << 28)
 #define TSTAMPS_EN(x)        ((x) << 29)
 #define SACK_EN(x)           ((x) << 30)
@@ -292,6 +312,9 @@ struct cpl_pass_establish {
 	union opcode_tid ot;
 	__be32 rsvd;
 	__be32 tos_stid;
+#define PASS_OPEN_TID(x) ((x) << 0)
+#define PASS_OPEN_TOS(x) ((x) << 24)
+#define GET_PASS_OPEN_TID(x)	(((x) >> 0) & 0xFFFFFF)
 #define GET_POPEN_TID(x) ((x) & 0xffffff)
 #define GET_POPEN_TOS(x) (((x) >> 24) & 0xff)
 	__be16 mac_idx;
@@ -332,6 +355,7 @@ struct cpl_set_tcb_field {
 	__be16 word_cookie;
 #define TCB_WORD(x)   ((x) << 0)
 #define TCB_COOKIE(x) ((x) << 5)
+#define GET_TCB_COOKIE(x) (((x) >> 5) & 7)
 	__be64 mask;
 	__be64 val;
 };
@@ -536,6 +560,37 @@ struct cpl_rx_pkt {
 	__be16 err_vec;
 };
 
+/* rx_pkt.l2info fields */
+#define S_RX_ETHHDR_LEN    0
+#define M_RX_ETHHDR_LEN    0x1F
+#define V_RX_ETHHDR_LEN(x) ((x) << S_RX_ETHHDR_LEN)
+#define G_RX_ETHHDR_LEN(x) (((x) >> S_RX_ETHHDR_LEN) & M_RX_ETHHDR_LEN)
+
+#define S_RX_MACIDX    8
+#define M_RX_MACIDX    0x1FF
+#define V_RX_MACIDX(x) ((x) << S_RX_MACIDX)
+#define G_RX_MACIDX(x) (((x) >> S_RX_MACIDX) & M_RX_MACIDX)
+
+#define S_RXF_SYN    21
+#define V_RXF_SYN(x) ((x) << S_RXF_SYN)
+#define F_RXF_SYN    V_RXF_SYN(1U)
+
+#define S_RX_CHAN    28
+#define M_RX_CHAN    0xF
+#define V_RX_CHAN(x) ((x) << S_RX_CHAN)
+#define G_RX_CHAN(x) (((x) >> S_RX_CHAN) & M_RX_CHAN)
+
+/* rx_pkt.hdr_len fields */
+#define S_RX_TCPHDR_LEN    0
+#define M_RX_TCPHDR_LEN    0x3F
+#define V_RX_TCPHDR_LEN(x) ((x) << S_RX_TCPHDR_LEN)
+#define G_RX_TCPHDR_LEN(x) (((x) >> S_RX_TCPHDR_LEN) & M_RX_TCPHDR_LEN)
+
+#define S_RX_IPHDR_LEN    6
+#define M_RX_IPHDR_LEN    0x3FF
+#define V_RX_IPHDR_LEN(x) ((x) << S_RX_IPHDR_LEN)
+#define G_RX_IPHDR_LEN(x) (((x) >> S_RX_IPHDR_LEN) & M_RX_IPHDR_LEN)
+
 struct cpl_trace_pkt {
 	u8 opcode;
 	u8 intf;
@@ -634,6 +689,17 @@ struct cpl_fw6_msg {
 /* cpl_fw6_msg.type values */
 enum {
 	FW6_TYPE_CMD_RPL = 0,
+	FW6_TYPE_WR_RPL = 1,
+	FW6_TYPE_CQE = 2,
+	FW6_TYPE_OFLD_CONNECTION_WR_RPL = 3,
+};
+
+struct cpl_fw6_msg_ofld_connection_wr_rpl {
+	__u64   cookie;
+	__be32  tid;    /* or atid in case of active failure */
+	__u8    t_state;
+	__u8    retval;
+	__u8    rsvd[2];
 };
 
 enum {

drivers/net/ethernet/chelsio/cxgb4/t4_regs.h

@@ -1064,4 +1064,41 @@
 #define  ADDRESS(x)     ((x) << ADDRESS_SHIFT)
 
 #define XGMAC_PORT_INT_CAUSE 0x10dc
+
+#define A_TP_TX_MOD_QUEUE_REQ_MAP 0x7e28
+
+#define A_TP_TX_MOD_CHANNEL_WEIGHT 0x7e34
+
+#define S_TX_MOD_QUEUE_REQ_MAP    0
+#define M_TX_MOD_QUEUE_REQ_MAP    0xffffU
+#define V_TX_MOD_QUEUE_REQ_MAP(x) ((x) << S_TX_MOD_QUEUE_REQ_MAP)
+
+#define A_TP_TX_MOD_QUEUE_WEIGHT0 0x7e30
+
+#define S_TX_MODQ_WEIGHT3    24
+#define M_TX_MODQ_WEIGHT3    0xffU
+#define V_TX_MODQ_WEIGHT3(x) ((x) << S_TX_MODQ_WEIGHT3)
+
+#define S_TX_MODQ_WEIGHT2    16
+#define M_TX_MODQ_WEIGHT2    0xffU
+#define V_TX_MODQ_WEIGHT2(x) ((x) << S_TX_MODQ_WEIGHT2)
+
+#define S_TX_MODQ_WEIGHT1    8
+#define M_TX_MODQ_WEIGHT1    0xffU
+#define V_TX_MODQ_WEIGHT1(x) ((x) << S_TX_MODQ_WEIGHT1)
+
+#define S_TX_MODQ_WEIGHT0    0
+#define M_TX_MODQ_WEIGHT0    0xffU
+#define V_TX_MODQ_WEIGHT0(x) ((x) << S_TX_MODQ_WEIGHT0)
+
+#define A_TP_TX_SCHED_HDR 0x23
+
+#define A_TP_TX_SCHED_FIFO 0x24
+
+#define A_TP_TX_SCHED_PCMD 0x25
+
+#define S_PORT    1
+#define V_PORT(x) ((x) << S_PORT)
+#define F_PORT    V_PORT(1U)
+
 #endif /* __T4_REGS_H */

drivers/net/ethernet/mellanox/mlx4/fw.c

@@ -1338,6 +1338,7 @@ int mlx4_QUERY_HCA(struct mlx4_dev *dev,
 {
 	struct mlx4_cmd_mailbox *mailbox;
 	__be32 *outbox;
+	u32 dword_field;
 	int err;
 	u8 byte_field;
 
@@ -1372,10 +1373,18 @@ int mlx4_QUERY_HCA(struct mlx4_dev *dev,
 	MLX4_GET(param->rdmarc_base,   outbox, INIT_HCA_RDMARC_BASE_OFFSET);
 	MLX4_GET(param->log_rd_per_qp, outbox, INIT_HCA_LOG_RD_OFFSET);
 
+	MLX4_GET(dword_field, outbox, INIT_HCA_FLAGS_OFFSET);
+	if (dword_field & (1 << INIT_HCA_DEVICE_MANAGED_FLOW_STEERING_EN)) {
+		param->steering_mode = MLX4_STEERING_MODE_DEVICE_MANAGED;
+	} else {
+		MLX4_GET(byte_field, outbox, INIT_HCA_UC_STEERING_OFFSET);
+		if (byte_field & 0x8)
+			param->steering_mode = MLX4_STEERING_MODE_B0;
+		else
+			param->steering_mode = MLX4_STEERING_MODE_A0;
+	}
 	/* steering attributes */
-	if (dev->caps.steering_mode ==
-	    MLX4_STEERING_MODE_DEVICE_MANAGED) {
-
+	if (param->steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED) {
 		MLX4_GET(param->mc_base, outbox, INIT_HCA_FS_BASE_OFFSET);
 		MLX4_GET(param->log_mc_entry_sz, outbox,
 			 INIT_HCA_FS_LOG_ENTRY_SZ_OFFSET);

drivers/net/ethernet/mellanox/mlx4/fw.h

@@ -172,6 +172,7 @@ struct mlx4_init_hca_param {
 	u8  log_uar_sz;
 	u8  uar_page_sz; /* log pg sz in 4k chunks */
 	u8  fs_hash_enable_bits;
+	u8  steering_mode; /* for QUERY_HCA */
 	u64 dev_cap_enabled;
 };
 

drivers/net/ethernet/mellanox/mlx4/main.c

@@ -85,15 +85,15 @@ static int probe_vf;
 module_param(probe_vf, int, 0644);
 MODULE_PARM_DESC(probe_vf, "number of vfs to probe by pf driver (num_vfs > 0)");
 
-int mlx4_log_num_mgm_entry_size = 10;
+int mlx4_log_num_mgm_entry_size = MLX4_DEFAULT_MGM_LOG_ENTRY_SIZE;
 module_param_named(log_num_mgm_entry_size,
 			mlx4_log_num_mgm_entry_size, int, 0444);
 MODULE_PARM_DESC(log_num_mgm_entry_size, "log mgm size, that defines the num"
 					 " of qp per mcg, for example:"
-					 " 10 gives 248.range: 9<="
+					 " 10 gives 248.range: 7 <="
 					 " log_num_mgm_entry_size <= 12."
-					 " Not in use with device managed"
-					 " flow steering");
+					 " To activate device managed"
+					 " flow steering when available, set to -1");
 
 static bool enable_64b_cqe_eqe;
 module_param(enable_64b_cqe_eqe, bool, 0444);
@@ -281,28 +281,6 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 	dev->caps.max_rss_tbl_sz     = dev_cap->max_rss_tbl_sz;
 
-	if (dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_FS_EN) {
-		dev->caps.steering_mode = MLX4_STEERING_MODE_DEVICE_MANAGED;
-		dev->caps.num_qp_per_mgm = dev_cap->fs_max_num_qp_per_entry;
-		dev->caps.fs_log_max_ucast_qp_range_size =
-			dev_cap->fs_log_max_ucast_qp_range_size;
-	} else {
-		if (dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_UC_STEER &&
-		    dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_MC_STEER) {
-			dev->caps.steering_mode = MLX4_STEERING_MODE_B0;
-		} else {
-			dev->caps.steering_mode = MLX4_STEERING_MODE_A0;
-
-			if (dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_UC_STEER ||
-			    dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_MC_STEER)
-				mlx4_warn(dev, "Must have UC_STEER and MC_STEER flags "
-						"set to use B0 steering. Falling back to A0 steering mode.\n");
-		}
-		dev->caps.num_qp_per_mgm = mlx4_get_qp_per_mgm(dev);
-	}
-	mlx4_dbg(dev, "Steering mode is: %s\n",
-		 mlx4_steering_mode_str(dev->caps.steering_mode));
-
 	/* Sense port always allowed on supported devices for ConnectX-1 and -2 */
 	if (mlx4_priv(dev)->pci_dev_data & MLX4_PCI_DEV_FORCE_SENSE_PORT)
 		dev->caps.flags |= MLX4_DEV_CAP_FLAG_SENSE_SUPPORT;
@@ -493,6 +471,23 @@ int mlx4_is_slave_active(struct mlx4_dev *dev, int slave)
 }
 EXPORT_SYMBOL(mlx4_is_slave_active);
 
+static void slave_adjust_steering_mode(struct mlx4_dev *dev,
+				       struct mlx4_dev_cap *dev_cap,
+				       struct mlx4_init_hca_param *hca_param)
+{
+	dev->caps.steering_mode = hca_param->steering_mode;
+	if (dev->caps.steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED) {
+		dev->caps.num_qp_per_mgm = dev_cap->fs_max_num_qp_per_entry;
+		dev->caps.fs_log_max_ucast_qp_range_size =
+			dev_cap->fs_log_max_ucast_qp_range_size;
+	} else
+		dev->caps.num_qp_per_mgm =
+			4 * ((1 << hca_param->log_mc_entry_sz)/16 - 2);
+
+	mlx4_dbg(dev, "Steering mode is: %s\n",
+		 mlx4_steering_mode_str(dev->caps.steering_mode));
+}
+
 static int mlx4_slave_cap(struct mlx4_dev *dev)
 {
 	int			   err;
@@ -635,6 +630,8 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 		dev->caps.cqe_size   = 32;
 	}
 
+	slave_adjust_steering_mode(dev, &dev_cap, &hca_param);
+
 	return 0;
 
 err_mem:
@@ -1321,6 +1318,59 @@ static void mlx4_parav_master_pf_caps(struct mlx4_dev *dev)
 	}
 }
 
+static int choose_log_fs_mgm_entry_size(int qp_per_entry)
+{
+	int i = MLX4_MIN_MGM_LOG_ENTRY_SIZE;
+
+	for (i = MLX4_MIN_MGM_LOG_ENTRY_SIZE; i <= MLX4_MAX_MGM_LOG_ENTRY_SIZE;
+	      i++) {
+		if (qp_per_entry <= 4 * ((1 << i) / 16 - 2))
+			break;
+	}
+
+	return (i <= MLX4_MAX_MGM_LOG_ENTRY_SIZE) ? i : -1;
+}
+
+static void choose_steering_mode(struct mlx4_dev *dev,
+				 struct mlx4_dev_cap *dev_cap)
+{
+	if (mlx4_log_num_mgm_entry_size == -1 &&
+	    dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_FS_EN &&
+	    (!mlx4_is_mfunc(dev) ||
+	     (dev_cap->fs_max_num_qp_per_entry >= (num_vfs + 1))) &&
+	    choose_log_fs_mgm_entry_size(dev_cap->fs_max_num_qp_per_entry) >=
+		MLX4_MIN_MGM_LOG_ENTRY_SIZE) {
+		dev->oper_log_mgm_entry_size =
+			choose_log_fs_mgm_entry_size(dev_cap->fs_max_num_qp_per_entry);
+		dev->caps.steering_mode = MLX4_STEERING_MODE_DEVICE_MANAGED;
+		dev->caps.num_qp_per_mgm = dev_cap->fs_max_num_qp_per_entry;
+		dev->caps.fs_log_max_ucast_qp_range_size =
+			dev_cap->fs_log_max_ucast_qp_range_size;
+	} else {
+		if (dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_UC_STEER &&
+		    dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_MC_STEER)
+			dev->caps.steering_mode = MLX4_STEERING_MODE_B0;
+		else {
+			dev->caps.steering_mode = MLX4_STEERING_MODE_A0;
+
+			if (dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_UC_STEER ||
+			    dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_MC_STEER)
+				mlx4_warn(dev, "Must have both UC_STEER and MC_STEER flags "
+					  "set to use B0 steering. Falling back to A0 steering mode.\n");
+		}
+		dev->oper_log_mgm_entry_size =
+			mlx4_log_num_mgm_entry_size > 0 ?
+			mlx4_log_num_mgm_entry_size :
+			MLX4_DEFAULT_MGM_LOG_ENTRY_SIZE;
+		dev->caps.num_qp_per_mgm = mlx4_get_qp_per_mgm(dev);
+	}
+	mlx4_dbg(dev, "Steering mode is: %s, oper_log_mgm_entry_size = %d, "
+		 "modparam log_num_mgm_entry_size = %d\n",
+		 mlx4_steering_mode_str(dev->caps.steering_mode),
+		 dev->oper_log_mgm_entry_size,
+		 mlx4_log_num_mgm_entry_size);
+}
+
 static int mlx4_init_hca(struct mlx4_dev *dev)
 {
 	struct mlx4_priv	  *priv = mlx4_priv(dev);
@@ -1360,6 +1410,8 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 			goto err_stop_fw;
 		}
 
+		choose_steering_mode(dev, &dev_cap);
+
 		if (mlx4_is_master(dev))
 			mlx4_parav_master_pf_caps(dev);
 
@@ -2452,6 +2504,17 @@ static int __init mlx4_verify_params(void)
 		port_type_array[0] = true;
 	}
 
+	if (mlx4_log_num_mgm_entry_size != -1 &&
+	    (mlx4_log_num_mgm_entry_size < MLX4_MIN_MGM_LOG_ENTRY_SIZE ||
+	     mlx4_log_num_mgm_entry_size > MLX4_MAX_MGM_LOG_ENTRY_SIZE)) {
+		pr_warning("mlx4_core: mlx4_log_num_mgm_entry_size (%d) not "
+			   "in legal range (-1 or %d..%d)\n",
+			   mlx4_log_num_mgm_entry_size,
+			   MLX4_MIN_MGM_LOG_ENTRY_SIZE,
+			   MLX4_MAX_MGM_LOG_ENTRY_SIZE);
+		return -1;
+	}
+
 	return 0;
 }
 

drivers/net/ethernet/mellanox/mlx4/mcg.c

@@ -54,12 +54,7 @@ struct mlx4_mgm {
 
 int mlx4_get_mgm_entry_size(struct mlx4_dev *dev)
 {
-	if (dev->caps.steering_mode ==
-	    MLX4_STEERING_MODE_DEVICE_MANAGED)
-		return 1 << MLX4_FS_MGM_LOG_ENTRY_SIZE;
-	else
-		return min((1 << mlx4_log_num_mgm_entry_size),
-			   MLX4_MAX_MGM_ENTRY_SIZE);
+	return 1 << dev->oper_log_mgm_entry_size;
 }
 
 int mlx4_get_qp_per_mgm(struct mlx4_dev *dev)

drivers/net/ethernet/mellanox/mlx4/mlx4.h

@@ -94,8 +94,10 @@ enum {
 };
 
 enum {
-	MLX4_MAX_MGM_ENTRY_SIZE = 0x1000,
-	MLX4_MAX_QP_PER_MGM	= 4 * (MLX4_MAX_MGM_ENTRY_SIZE / 16 - 2),
+	MLX4_DEFAULT_MGM_LOG_ENTRY_SIZE = 10,
+	MLX4_MIN_MGM_LOG_ENTRY_SIZE = 7,
+	MLX4_MAX_MGM_LOG_ENTRY_SIZE = 12,
+	MLX4_MAX_QP_PER_MGM = 4 * ((1 << MLX4_MAX_MGM_LOG_ENTRY_SIZE) / 16 - 2),
 	MLX4_MTT_ENTRY_PER_SEG	= 8,
 };
 

drivers/net/ethernet/mellanox/mlx4/resource_tracker.c

@@ -3071,6 +3071,7 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
 	struct list_head *rlist = &tracker->slave_list[slave].res_list[RES_MAC];
 	int err;
+	int qpn;
 	struct mlx4_net_trans_rule_hw_ctrl *ctrl;
 	struct _rule_hw  *rule_header;
 	int header_id;
@@ -3080,13 +3081,21 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 		return -EOPNOTSUPP;
 
 	ctrl = (struct mlx4_net_trans_rule_hw_ctrl *)inbox->buf;
+	qpn = be32_to_cpu(ctrl->qpn) & 0xffffff;
+	err = get_res(dev, slave, qpn, RES_QP, NULL);
+	if (err) {
+		pr_err("Steering rule with qpn 0x%x rejected.\n", qpn);
+		return err;
+	}
 	rule_header = (struct _rule_hw *)(ctrl + 1);
 	header_id = map_hw_to_sw_id(be16_to_cpu(rule_header->id));
 
 	switch (header_id) {
 	case MLX4_NET_TRANS_RULE_ID_ETH:
-		if (validate_eth_header_mac(slave, rule_header, rlist))
-			return -EINVAL;
+		if (validate_eth_header_mac(slave, rule_header, rlist)) {
+			err = -EINVAL;
+			goto err_put;
+		}
 		break;
 	case MLX4_NET_TRANS_RULE_ID_IB:
 		break;
@@ -3094,14 +3103,17 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 	case MLX4_NET_TRANS_RULE_ID_TCP:
 	case MLX4_NET_TRANS_RULE_ID_UDP:
 		pr_warn("Can't attach FS rule without L2 headers, adding L2 header.\n");
-		if (add_eth_header(dev, slave, inbox, rlist, header_id))
-			return -EINVAL;
+		if (add_eth_header(dev, slave, inbox, rlist, header_id)) {
+			err = -EINVAL;
+			goto err_put;
+		}
 		vhcr->in_modifier +=
 			sizeof(struct mlx4_net_trans_rule_hw_eth) >> 2;
 		break;
 	default:
 		pr_err("Corrupted mailbox.\n");
-		return -EINVAL;
+		err = -EINVAL;
+		goto err_put;
 	}
 
 	err = mlx4_cmd_imm(dev, inbox->dma, &vhcr->out_param,
@@ -3109,16 +3121,18 @@ int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 			   MLX4_QP_FLOW_STEERING_ATTACH, MLX4_CMD_TIME_CLASS_A,
 			   MLX4_CMD_NATIVE);
 	if (err)
-		return err;
+		goto err_put;
 
 	err = add_res_range(dev, slave, vhcr->out_param, 1, RES_FS_RULE, 0);
 	if (err) {
 		mlx4_err(dev, "Fail to add flow steering resources.\n ");
 		/* detach rule*/
 		mlx4_cmd(dev, vhcr->out_param, 0, 0,
-			 MLX4_QP_FLOW_STEERING_ATTACH, MLX4_CMD_TIME_CLASS_A,
+			 MLX4_QP_FLOW_STEERING_DETACH, MLX4_CMD_TIME_CLASS_A,
 			 MLX4_CMD_NATIVE);
 	}
+err_put:
+	put_res(dev, slave, qpn, RES_QP);
 	return err;
 }
 

drivers/scsi/csiostor/t4fw_api_stor.h

@@ -40,45 +40,6 @@
  *   R E T U R N   V A L U E S
  ********************************/
 
-enum fw_retval {
-	FW_SUCCESS		= 0,	/* completed sucessfully */
-	FW_EPERM		= 1,	/* operation not permitted */
-	FW_ENOENT		= 2,	/* no such file or directory */
-	FW_EIO			= 5,	/* input/output error; hw bad */
-	FW_ENOEXEC		= 8,	/* exec format error; inv microcode */
-	FW_EAGAIN		= 11,	/* try again */
-	FW_ENOMEM		= 12,	/* out of memory */
-	FW_EFAULT		= 14,	/* bad address; fw bad */
-	FW_EBUSY		= 16,	/* resource busy */
-	FW_EEXIST		= 17,	/* file exists */
-	FW_EINVAL		= 22,	/* invalid argument */
-	FW_ENOSPC		= 28,	/* no space left on device */
-	FW_ENOSYS		= 38,	/* functionality not implemented */
-	FW_EPROTO		= 71,	/* protocol error */
-	FW_EADDRINUSE		= 98,	/* address already in use */
-	FW_EADDRNOTAVAIL	= 99,	/* cannot assigned requested address */
-	FW_ENETDOWN		= 100,	/* network is down */
-	FW_ENETUNREACH		= 101,	/* network is unreachable */
-	FW_ENOBUFS		= 105,	/* no buffer space available */
-	FW_ETIMEDOUT		= 110,	/* timeout */
-	FW_EINPROGRESS		= 115,	/* fw internal */
-	FW_SCSI_ABORT_REQUESTED	= 128,	/* */
-	FW_SCSI_ABORT_TIMEDOUT	= 129,	/* */
-	FW_SCSI_ABORTED		= 130,	/* */
-	FW_SCSI_CLOSE_REQUESTED	= 131,	/* */
-	FW_ERR_LINK_DOWN	= 132,	/* */
-	FW_RDEV_NOT_READY	= 133,	/* */
-	FW_ERR_RDEV_LOST	= 134,	/* */
-	FW_ERR_RDEV_LOGO	= 135,	/* */
-	FW_FCOE_NO_XCHG		= 136,	/* */
-	FW_SCSI_RSP_ERR		= 137,	/* */
-	FW_ERR_RDEV_IMPL_LOGO	= 138,	/* */
-	FW_SCSI_UNDER_FLOW_ERR  = 139,	/* */
-	FW_SCSI_OVER_FLOW_ERR   = 140,	/* */
-	FW_SCSI_DDP_ERR		= 141,	/* DDP error*/
-	FW_SCSI_TASK_ERR	= 142,	/* No SCSI tasks available */
-};
-
 enum fw_fcoe_link_sub_op {
 	FCOE_LINK_DOWN	= 0x0,
 	FCOE_LINK_UP	= 0x1,

include/linux/mlx4/device.h

@@ -625,6 +625,7 @@ struct mlx4_dev {
 	u8			rev_id;
 	char			board_id[MLX4_BOARD_ID_LEN];
 	int			num_vfs;
+	int			oper_log_mgm_entry_size;
 	u64			regid_promisc_array[MLX4_MAX_PORTS + 1];
 	u64			regid_allmulti_array[MLX4_MAX_PORTS + 1];
 };