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

Pull networking updates from David Miller:
 "Another merge window, another set of networking changes.  I've heard
  rumblings that the lightweight tunnels infrastructure has been voted
  networking change of the year.  But what do I know?

   1) Add conntrack support to openvswitch, from Joe Stringer.

   2) Initial support for VRF (Virtual Routing and Forwarding), which
      allows the segmentation of routing paths without using multiple
      devices.  There are some semantic kinks to work out still, but
      this is a reasonably strong foundation.  From David Ahern.

   3) Remove spinlock fro act_bpf fast path, from Alexei Starovoitov.

   4) Ignore route nexthops with a link down state in ipv6, just like
      ipv4.  From Andy Gospodarek.

   5) Remove spinlock from fast path of act_gact and act_mirred, from
      Eric Dumazet.

   6) Document the DSA layer, from Florian Fainelli.

   7) Add netconsole support to bcmgenet, systemport, and DSA.  Also
      from Florian Fainelli.

   8) Add Mellanox Switch Driver and core infrastructure, from Jiri
      Pirko.

   9) Add support for "light weight tunnels", which allow for
      encapsulation and decapsulation without bearing the overhead of a
      full blown netdevice.  From Thomas Graf, Jiri Benc, and a cast of
      others.

  10) Add Identifier Locator Addressing support for ipv6, from Tom
      Herbert.

  11) Support fragmented SKBs in iwlwifi, from Johannes Berg.

  12) Allow perf PMUs to be accessed from eBPF programs, from Kaixu Xia.

  13) Add BQL support to 3c59x driver, from Loganaden Velvindron.

  14) Stop using a zero TX queue length to mean that a device shouldn't
      have a qdisc attached, use an explicit flag instead.  From Phil
      Sutter.

  15) Use generic geneve netdevice infrastructure in openvswitch, from
      Pravin B Shelar.

  16) Add infrastructure to avoid re-forwarding a packet in software
      that was already forwarded by a hardware switch.  From Scott
      Feldman.

  17) Allow AF_PACKET fanout function to be implemented in a bpf
      program, from Willem de Bruijn"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1458 commits)
  netfilter: nf_conntrack: make nf_ct_zone_dflt built-in
  netfilter: nf_dup{4, 6}: fix build error when nf_conntrack disabled
  net: fec: clear receive interrupts before processing a packet
  ipv6: fix exthdrs offload registration in out_rt path
  xen-netback: add support for multicast control
  bgmac: Update fixed_phy_register()
  sock, diag: fix panic in sock_diag_put_filterinfo
  flow_dissector: Use 'const' where possible.
  flow_dissector: Fix function argument ordering dependency
  ixgbe: Resolve "initialized field overwritten" warnings
  ixgbe: Remove bimodal SR-IOV disabling
  ixgbe: Add support for reporting 2.5G link speed
  ixgbe: fix bounds checking in ixgbe_setup_tc for 82598
  ixgbe: support for ethtool set_rxfh
  ixgbe: Avoid needless PHY access on copper phys
  ixgbe: cleanup to use cached mask value
  ixgbe: Remove second instance of lan_id variable
  ixgbe: use kzalloc for allocating one thing
  flow: Move __get_hash_from_flowi{4,6} into flow_dissector.c
  ixgbe: Remove unused PCI bus types
  ...

17 files changed, 1507 insertions, 175 deletions

diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
index 629f75d70353..fa0c7b54ec7a 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
@@ -767,6 +767,11 @@ struct adapter {
 	bool tid_release_task_busy;
 
 	struct dentry *debugfs_root;
+	u32 use_bd;     /* Use SGE Back Door intfc for reading SGE Contexts */
+	u32 trace_rss;	/* 1 implies that different RSS flit per filter is
+			 * used per filter else if 0 default RSS flit is
+			 * used for all 4 filters.
+			 */
 
 	spinlock_t stats_lock;
 	spinlock_t win0_lock ____cacheline_aligned_in_smp;
@@ -1284,6 +1289,7 @@ int t4_fwcache(struct adapter *adap, enum fw_params_param_dev_fwcache op);
 int t4_fw_upgrade(struct adapter *adap, unsigned int mbox,
 		  const u8 *fw_data, unsigned int size, int force);
 unsigned int t4_flash_cfg_addr(struct adapter *adapter);
+int t4_check_fw_version(struct adapter *adap);
 int t4_get_fw_version(struct adapter *adapter, u32 *vers);
 int t4_get_tp_version(struct adapter *adapter, u32 *vers);
 int t4_get_exprom_version(struct adapter *adapter, u32 *vers);
@@ -1440,6 +1446,10 @@ int t4_sge_ctxt_flush(struct adapter *adap, unsigned int mbox);
 int t4_handle_fw_rpl(struct adapter *adap, const __be64 *rpl);
 void t4_db_full(struct adapter *adapter);
 void t4_db_dropped(struct adapter *adapter);
+int t4_set_trace_filter(struct adapter *adapter, const struct trace_params *tp,
+			int filter_index, int enable);
+void t4_get_trace_filter(struct adapter *adapter, struct trace_params *tp,
+			 int filter_index, int *enabled);
 int t4_fwaddrspace_write(struct adapter *adap, unsigned int mbox,
 			 u32 addr, u32 val);
 void t4_sge_decode_idma_state(struct adapter *adapter, int state);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c
index 6074680bc985..052c660aca80 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c
@@ -31,6 +31,15 @@ static const char * const dcb_ver_array[] = {
 	"Auto Negotiated"
 };
 
+static inline bool cxgb4_dcb_state_synced(enum cxgb4_dcb_state state)
+{
+	if (state == CXGB4_DCB_STATE_FW_ALLSYNCED ||
+	    state == CXGB4_DCB_STATE_HOST)
+		return true;
+	else
+		return false;
+}
+
 /* Initialize a port's Data Center Bridging state.  Typically used after a
  * Link Down event.
  */
@@ -603,7 +612,7 @@ static void cxgb4_getpfccfg(struct net_device *dev, int priority, u8 *pfccfg)
 	struct port_info *pi = netdev2pinfo(dev);
 	struct port_dcb_info *dcb = &pi->dcb;
 
-	if (dcb->state != CXGB4_DCB_STATE_FW_ALLSYNCED ||
+	if (!cxgb4_dcb_state_synced(dcb->state) ||
 	    priority >= CXGB4_MAX_PRIORITY)
 		*pfccfg = 0;
 	else
@@ -620,7 +629,7 @@ static void cxgb4_setpfccfg(struct net_device *dev, int priority, u8 pfccfg)
 	struct adapter *adap = pi->adapter;
 	int err;
 
-	if (pi->dcb.state != CXGB4_DCB_STATE_FW_ALLSYNCED ||
+	if (!cxgb4_dcb_state_synced(pi->dcb.state) ||
 	    priority >= CXGB4_MAX_PRIORITY)
 		return;
 
@@ -732,7 +741,7 @@ static u8 cxgb4_getpfcstate(struct net_device *dev)
 {
 	struct port_info *pi = netdev2pinfo(dev);
 
-	if (pi->dcb.state != CXGB4_DCB_STATE_FW_ALLSYNCED)
+	if (!cxgb4_dcb_state_synced(pi->dcb.state))
 		return false;
 
 	return pi->dcb.pfcen != 0;
@@ -756,7 +765,7 @@ static int __cxgb4_getapp(struct net_device *dev, u8 app_idtype, u16 app_id,
 	struct adapter *adap = pi->adapter;
 	int i;
 
-	if (pi->dcb.state != CXGB4_DCB_STATE_FW_ALLSYNCED)
+	if (!cxgb4_dcb_state_synced(pi->dcb.state))
 		return 0;
 
 	for (i = 0; i < CXGB4_MAX_DCBX_APP_SUPPORTED; i++) {
@@ -794,7 +803,9 @@ static int __cxgb4_getapp(struct net_device *dev, u8 app_idtype, u16 app_id,
  */
 static int cxgb4_getapp(struct net_device *dev, u8 app_idtype, u16 app_id)
 {
-	return __cxgb4_getapp(dev, app_idtype, app_id, 0);
+	/* Convert app_idtype to firmware format before querying */
+	return __cxgb4_getapp(dev, app_idtype == DCB_APP_IDTYPE_ETHTYPE ?
+			      app_idtype : 3, app_id, 0);
 }
 
 /* Write a new Application User Priority Map for the specified Application ID
@@ -808,7 +819,7 @@ static int __cxgb4_setapp(struct net_device *dev, u8 app_idtype, u16 app_id,
 	int i, err;
 
 
-	if (pi->dcb.state != CXGB4_DCB_STATE_FW_ALLSYNCED)
+	if (!cxgb4_dcb_state_synced(pi->dcb.state))
 		return -EINVAL;
 
 	/* DCB info gets thrown away on link up */
@@ -896,10 +907,11 @@ cxgb4_ieee_negotiation_complete(struct net_device *dev,
 	struct port_info *pi = netdev2pinfo(dev);
 	struct port_dcb_info *dcb = &pi->dcb;
 
-	if (dcb_subtype && !(dcb->msgs & dcb_subtype))
-		return 0;
+	if (dcb->state == CXGB4_DCB_STATE_FW_ALLSYNCED)
+		if (dcb_subtype && !(dcb->msgs & dcb_subtype))
+			return 0;
 
-	return (dcb->state == CXGB4_DCB_STATE_FW_ALLSYNCED &&
+	return (cxgb4_dcb_state_synced(dcb->state) &&
 		(dcb->supported & DCB_CAP_DCBX_VER_IEEE));
 }
 
@@ -1057,7 +1069,7 @@ static u8 cxgb4_setdcbx(struct net_device *dev, u8 dcb_request)
 
 	/* Can't enable DCB if we haven't successfully negotiated it.
 	 */
-	if (pi->dcb.state != CXGB4_DCB_STATE_FW_ALLSYNCED)
+	if (!cxgb4_dcb_state_synced(pi->dcb.state))
 		return 1;
 
 	/* There's currently no mechanism to allow for the firmware DCBX
@@ -1080,7 +1092,7 @@ static int cxgb4_getpeer_app(struct net_device *dev,
 	struct adapter *adap = pi->adapter;
 	int i, err = 0;
 
-	if (pi->dcb.state != CXGB4_DCB_STATE_FW_ALLSYNCED)
+	if (!cxgb4_dcb_state_synced(pi->dcb.state))
 		return 1;
 
 	info->willing = 0;
@@ -1114,7 +1126,7 @@ static int cxgb4_getpeerapp_tbl(struct net_device *dev, struct dcb_app *table)
 	struct adapter *adap = pi->adapter;
 	int i, err = 0;
 
-	if (pi->dcb.state != CXGB4_DCB_STATE_FW_ALLSYNCED)
+	if (!cxgb4_dcb_state_synced(pi->dcb.state))
 		return 1;
 
 	for (i = 0; i < CXGB4_MAX_DCBX_APP_SUPPORTED; i++) {
@@ -1133,7 +1145,7 @@ static int cxgb4_getpeerapp_tbl(struct net_device *dev, struct dcb_app *table)
 		if (!pcmd.u.dcb.app_priority.protocolid)
 			break;
 
-		table[i].selector = pcmd.u.dcb.app_priority.sel_field;
+		table[i].selector = (pcmd.u.dcb.app_priority.sel_field + 1);
 		table[i].protocol =
 			be16_to_cpu(pcmd.u.dcb.app_priority.protocolid);
 		table[i].priority =
@@ -1181,6 +1193,8 @@ static int cxgb4_cee_peer_getpg(struct net_device *dev, struct cee_pg *pg)
 	for (i = 0; i < CXGB4_MAX_PRIORITY; i++)
 		pg->pg_bw[i] = pcmd.u.dcb.pgrate.pgrate[i];
 
+	pg->tcs_supported = pcmd.u.dcb.pgrate.num_tcs_supported;
+
 	return 0;
 }
 
@@ -1198,6 +1212,8 @@ static int cxgb4_cee_peer_getpfc(struct net_device *dev, struct cee_pfc *pfc)
 	 */
 	pfc->pfc_en = bitswap_1(pi->dcb.pfcen);
 
+	pfc->tcs_supported = pi->dcb.pfc_num_tcs_supported;
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
index c3c7db41819d..0a87a3247464 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
@@ -151,6 +151,45 @@ static int cim_la_show_3in1(struct seq_file *seq, void *v, int idx)
 	return 0;
 }
 
+static int cim_la_show_t6(struct seq_file *seq, void *v, int idx)
+{
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(seq, "Status   Inst    Data      PC     LS0Stat  "
+			 "LS0Addr  LS0Data  LS1Stat  LS1Addr  LS1Data\n");
+	} else {
+		const u32 *p = v;
+
+		seq_printf(seq, "  %02x   %04x%04x %04x%04x %04x%04x %08x %08x %08x %08x %08x %08x\n",
+			   (p[9] >> 16) & 0xff,       /* Status */
+			   p[9] & 0xffff, p[8] >> 16, /* Inst */
+			   p[8] & 0xffff, p[7] >> 16, /* Data */
+			   p[7] & 0xffff, p[6] >> 16, /* PC */
+			   p[2], p[1], p[0],      /* LS0 Stat, Addr and Data */
+			   p[5], p[4], p[3]);     /* LS1 Stat, Addr and Data */
+	}
+	return 0;
+}
+
+static int cim_la_show_pc_t6(struct seq_file *seq, void *v, int idx)
+{
+	if (v == SEQ_START_TOKEN) {
+		seq_puts(seq, "Status   Inst    Data      PC\n");
+	} else {
+		const u32 *p = v;
+
+		seq_printf(seq, "  %02x   %08x %08x %08x\n",
+			   p[3] & 0xff, p[2], p[1], p[0]);
+		seq_printf(seq, "  %02x   %02x%06x %02x%06x %02x%06x\n",
+			   (p[6] >> 8) & 0xff, p[6] & 0xff, p[5] >> 8,
+			   p[5] & 0xff, p[4] >> 8, p[4] & 0xff, p[3] >> 8);
+		seq_printf(seq, "  %02x   %04x%04x %04x%04x %04x%04x\n",
+			   (p[9] >> 16) & 0xff, p[9] & 0xffff, p[8] >> 16,
+			   p[8] & 0xffff, p[7] >> 16, p[7] & 0xffff,
+			   p[6] >> 16);
+	}
+	return 0;
+}
+
 static int cim_la_open(struct inode *inode, struct file *file)
 {
 	int ret;
@@ -162,9 +201,18 @@ static int cim_la_open(struct inode *inode, struct file *file)
 	if (ret)
 		return ret;
 
-	p = seq_open_tab(file, adap->params.cim_la_size / 8, 8 * sizeof(u32), 1,
-			 cfg & UPDBGLACAPTPCONLY_F ?
-			 cim_la_show_3in1 : cim_la_show);
+	if (is_t6(adap->params.chip)) {
+		/* +1 to account for integer division of CIMLA_SIZE/10 */
+		p = seq_open_tab(file, (adap->params.cim_la_size / 10) + 1,
+				 10 * sizeof(u32), 1,
+				 cfg & UPDBGLACAPTPCONLY_F ?
+					cim_la_show_pc_t6 : cim_la_show_t6);
+	} else {
+		p = seq_open_tab(file, adap->params.cim_la_size / 8,
+				 8 * sizeof(u32), 1,
+				 cfg & UPDBGLACAPTPCONLY_F ? cim_la_show_3in1 :
+							     cim_la_show);
+	}
 	if (!p)
 		return -ENOMEM;
 
@@ -298,11 +346,11 @@ static int cim_qcfg_show(struct seq_file *seq, void *v)
 		if (is_t4(adap->params.chip)) {
 			i = t4_cim_read(adap, UP_OBQ_0_REALADDR_A,
 					ARRAY_SIZE(obq_wr_t4), obq_wr_t4);
-				wr = obq_wr_t4;
+			wr = obq_wr_t4;
 		} else {
 			i = t4_cim_read(adap, UP_OBQ_0_SHADOW_REALADDR_A,
 					ARRAY_SIZE(obq_wr_t5), obq_wr_t5);
-				wr = obq_wr_t5;
+			wr = obq_wr_t5;
 		}
 	}
 	if (i)
@@ -1153,6 +1201,299 @@ static const struct file_operations mbox_debugfs_fops = {
 	.write   = mbox_write
 };
 
+static int mps_trc_show(struct seq_file *seq, void *v)
+{
+	int enabled, i;
+	struct trace_params tp;
+	unsigned int trcidx = (uintptr_t)seq->private & 3;
+	struct adapter *adap = seq->private - trcidx;
+
+	t4_get_trace_filter(adap, &tp, trcidx, &enabled);
+	if (!enabled) {
+		seq_puts(seq, "tracer is disabled\n");
+		return 0;
+	}
+
+	if (tp.skip_ofst * 8 >= TRACE_LEN) {
+		dev_err(adap->pdev_dev, "illegal trace pattern skip offset\n");
+		return -EINVAL;
+	}
+	if (tp.port < 8) {
+		i = adap->chan_map[tp.port & 3];
+		if (i >= MAX_NPORTS) {
+			dev_err(adap->pdev_dev, "tracer %u is assigned "
+				"to non-existing port\n", trcidx);
+			return -EINVAL;
+		}
+		seq_printf(seq, "tracer is capturing %s %s, ",
+			   adap->port[i]->name, tp.port < 4 ? "Rx" : "Tx");
+	} else
+		seq_printf(seq, "tracer is capturing loopback %d, ",
+			   tp.port - 8);
+	seq_printf(seq, "snap length: %u, min length: %u\n", tp.snap_len,
+		   tp.min_len);
+	seq_printf(seq, "packets captured %smatch filter\n",
+		   tp.invert ? "do not " : "");
+
+	if (tp.skip_ofst) {
+		seq_puts(seq, "filter pattern: ");
+		for (i = 0; i < tp.skip_ofst * 2; i += 2)
+			seq_printf(seq, "%08x%08x", tp.data[i], tp.data[i + 1]);
+		seq_putc(seq, '/');
+		for (i = 0; i < tp.skip_ofst * 2; i += 2)
+			seq_printf(seq, "%08x%08x", tp.mask[i], tp.mask[i + 1]);
+		seq_puts(seq, "@0\n");
+	}
+
+	seq_puts(seq, "filter pattern: ");
+	for (i = tp.skip_ofst * 2; i < TRACE_LEN / 4; i += 2)
+		seq_printf(seq, "%08x%08x", tp.data[i], tp.data[i + 1]);
+	seq_putc(seq, '/');
+	for (i = tp.skip_ofst * 2; i < TRACE_LEN / 4; i += 2)
+		seq_printf(seq, "%08x%08x", tp.mask[i], tp.mask[i + 1]);
+	seq_printf(seq, "@%u\n", (tp.skip_ofst + tp.skip_len) * 8);
+	return 0;
+}
+
+static int mps_trc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, mps_trc_show, inode->i_private);
+}
+
+static unsigned int xdigit2int(unsigned char c)
+{
+	return isdigit(c) ? c - '0' : tolower(c) - 'a' + 10;
+}
+
+#define TRC_PORT_NONE 0xff
+#define TRC_RSS_ENABLE 0x33
+#define TRC_RSS_DISABLE 0x13
+
+/* Set an MPS trace filter.  Syntax is:
+ *
+ * disable
+ *
+ * to disable tracing, or
+ *
+ * interface qid=<qid no> [snaplen=<val>] [minlen=<val>] [not] [<pattern>]...
+ *
+ * where interface is one of rxN, txN, or loopbackN, N = 0..3, qid can be one
+ * of the NIC's response qid obtained from sge_qinfo and pattern has the form
+ *
+ * <pattern data>[/<pattern mask>][@<anchor>]
+ *
+ * Up to 2 filter patterns can be specified.  If 2 are supplied the first one
+ * must be anchored at 0.  An omited mask is taken as a mask of 1s, an omitted
+ * anchor is taken as 0.
+ */
+static ssize_t mps_trc_write(struct file *file, const char __user *buf,
+			     size_t count, loff_t *pos)
+{
+	int i, enable, ret;
+	u32 *data, *mask;
+	struct trace_params tp;
+	const struct inode *ino;
+	unsigned int trcidx;
+	char *s, *p, *word, *end;
+	struct adapter *adap;
+	u32 j;
+
+	ino = file_inode(file);
+	trcidx = (uintptr_t)ino->i_private & 3;
+	adap = ino->i_private - trcidx;
+
+	/* Don't accept input more than 1K, can't be anything valid except lots
+	 * of whitespace.  Well, use less.
+	 */
+	if (count > 1024)
+		return -EFBIG;
+	p = s = kzalloc(count + 1, GFP_USER);
+	if (!s)
+		return -ENOMEM;
+	if (copy_from_user(s, buf, count)) {
+		count = -EFAULT;
+		goto out;
+	}
+
+	if (s[count - 1] == '\n')
+		s[count - 1] = '\0';
+
+	enable = strcmp("disable", s) != 0;
+	if (!enable)
+		goto apply;
+
+	/* enable or disable trace multi rss filter */
+	if (adap->trace_rss)
+		t4_write_reg(adap, MPS_TRC_CFG_A, TRC_RSS_ENABLE);
+	else
+		t4_write_reg(adap, MPS_TRC_CFG_A, TRC_RSS_DISABLE);
+
+	memset(&tp, 0, sizeof(tp));
+	tp.port = TRC_PORT_NONE;
+	i = 0;	/* counts pattern nibbles */
+
+	while (p) {
+		while (isspace(*p))
+			p++;
+		word = strsep(&p, " ");
+		if (!*word)
+			break;
+
+		if (!strncmp(word, "qid=", 4)) {
+			end = (char *)word + 4;
+			ret = kstrtouint(end, 10, &j);
+			if (ret)
+				goto out;
+			if (!adap->trace_rss) {
+				t4_write_reg(adap, MPS_T5_TRC_RSS_CONTROL_A, j);
+				continue;
+			}
+
+			switch (trcidx) {
+			case 0:
+				t4_write_reg(adap, MPS_TRC_RSS_CONTROL_A, j);
+				break;
+			case 1:
+				t4_write_reg(adap,
+					     MPS_TRC_FILTER1_RSS_CONTROL_A, j);
+				break;
+			case 2:
+				t4_write_reg(adap,
+					     MPS_TRC_FILTER2_RSS_CONTROL_A, j);
+				break;
+			case 3:
+				t4_write_reg(adap,
+					     MPS_TRC_FILTER3_RSS_CONTROL_A, j);
+				break;
+			}
+			continue;
+		}
+		if (!strncmp(word, "snaplen=", 8)) {
+			end = (char *)word + 8;
+			ret = kstrtouint(end, 10, &j);
+			if (ret || j > 9600) {
+inval:				count = -EINVAL;
+				goto out;
+			}
+			tp.snap_len = j;
+			continue;
+		}
+		if (!strncmp(word, "minlen=", 7)) {
+			end = (char *)word + 7;
+			ret = kstrtouint(end, 10, &j);
+			if (ret || j > TFMINPKTSIZE_M)
+				goto inval;
+			tp.min_len = j;
+			continue;
+		}
+		if (!strcmp(word, "not")) {
+			tp.invert = !tp.invert;
+			continue;
+		}
+		if (!strncmp(word, "loopback", 8) && tp.port == TRC_PORT_NONE) {
+			if (word[8] < '0' || word[8] > '3' || word[9])
+				goto inval;
+			tp.port = word[8] - '0' + 8;
+			continue;
+		}
+		if (!strncmp(word, "tx", 2) && tp.port == TRC_PORT_NONE) {
+			if (word[2] < '0' || word[2] > '3' || word[3])
+				goto inval;
+			tp.port = word[2] - '0' + 4;
+			if (adap->chan_map[tp.port & 3] >= MAX_NPORTS)
+				goto inval;
+			continue;
+		}
+		if (!strncmp(word, "rx", 2) && tp.port == TRC_PORT_NONE) {
+			if (word[2] < '0' || word[2] > '3' || word[3])
+				goto inval;
+			tp.port = word[2] - '0';
+			if (adap->chan_map[tp.port] >= MAX_NPORTS)
+				goto inval;
+			continue;
+		}
+		if (!isxdigit(*word))
+			goto inval;
+
+		/* we have found a trace pattern */
+		if (i) {                            /* split pattern */
+			if (tp.skip_len)            /* too many splits */
+				goto inval;
+			tp.skip_ofst = i / 16;
+		}
+
+		data = &tp.data[i / 8];
+		mask = &tp.mask[i / 8];
+		j = i;
+
+		while (isxdigit(*word)) {
+			if (i >= TRACE_LEN * 2) {
+				count = -EFBIG;
+				goto out;
+			}
+			*data = (*data << 4) + xdigit2int(*word++);
+			if (++i % 8 == 0)
+				data++;
+		}
+		if (*word == '/') {
+			word++;
+			while (isxdigit(*word)) {
+				if (j >= i)         /* mask longer than data */
+					goto inval;
+				*mask = (*mask << 4) + xdigit2int(*word++);
+				if (++j % 8 == 0)
+					mask++;
+			}
+			if (i != j)                 /* mask shorter than data */
+				goto inval;
+		} else {                            /* no mask, use all 1s */
+			for ( ; i - j >= 8; j += 8)
+				*mask++ = 0xffffffff;
+			if (i % 8)
+				*mask = (1 << (i % 8) * 4) - 1;
+		}
+		if (*word == '@') {
+			end = (char *)word + 1;
+			ret = kstrtouint(end, 10, &j);
+			if (*end && *end != '\n')
+				goto inval;
+			if (j & 7)          /* doesn't start at multiple of 8 */
+				goto inval;
+			j /= 8;
+			if (j < tp.skip_ofst)     /* overlaps earlier pattern */
+				goto inval;
+			if (j - tp.skip_ofst > 31)            /* skip too big */
+				goto inval;
+			tp.skip_len = j - tp.skip_ofst;
+		}
+		if (i % 8) {
+			*data <<= (8 - i % 8) * 4;
+			*mask <<= (8 - i % 8) * 4;
+			i = (i + 15) & ~15;         /* 8-byte align */
+		}
+	}
+
+	if (tp.port == TRC_PORT_NONE)
+		goto inval;
+
+apply:
+	i = t4_set_trace_filter(adap, &tp, trcidx, enable);
+	if (i)
+		count = i;
+out:
+	kfree(s);
+	return count;
+}
+
+static const struct file_operations mps_trc_debugfs_fops = {
+	.owner   = THIS_MODULE,
+	.open    = mps_trc_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+	.write   = mps_trc_write
+};
+
 static ssize_t flash_read(struct file *file, char __user *buf, size_t count,
 			  loff_t *ppos)
 {
@@ -1895,13 +2236,13 @@ static int sge_qinfo_show(struct seq_file *seq, void *v)
 {
 	struct adapter *adap = seq->private;
 	int eth_entries = DIV_ROUND_UP(adap->sge.ethqsets, 4);
-	int toe_entries = DIV_ROUND_UP(adap->sge.ofldqsets, 4);
+	int iscsi_entries = DIV_ROUND_UP(adap->sge.ofldqsets, 4);
 	int rdma_entries = DIV_ROUND_UP(adap->sge.rdmaqs, 4);
 	int ciq_entries = DIV_ROUND_UP(adap->sge.rdmaciqs, 4);
 	int ctrl_entries = DIV_ROUND_UP(MAX_CTRL_QUEUES, 4);
 	int i, r = (uintptr_t)v - 1;
-	int toe_idx = r - eth_entries;
-	int rdma_idx = toe_idx - toe_entries;
+	int iscsi_idx = r - eth_entries;
+	int rdma_idx = iscsi_idx - iscsi_entries;
 	int ciq_idx = rdma_idx - rdma_entries;
 	int ctrl_idx =  ciq_idx - ciq_entries;
 	int fq_idx =  ctrl_idx - ctrl_entries;
@@ -1917,8 +2258,12 @@ do { \
 		seq_putc(seq, '\n'); \
 } while (0)
 #define S(s, v) S3("s", s, v)
+#define T3(fmt_spec, s, v) S3(fmt_spec, s, tx[i].v)
 #define T(s, v) S3("u", s, tx[i].v)
+#define TL(s, v) T3("lu", s, v)
+#define R3(fmt_spec, s, v) S3(fmt_spec, s, rx[i].v)
 #define R(s, v) S3("u", s, rx[i].v)
+#define RL(s, v) R3("lu", s, v)
 
 	if (r < eth_entries) {
 		int base_qset = r * 4;
@@ -1957,12 +2302,30 @@ do { \
 		R("FL avail:", fl.avail);
 		R("FL PIDX:", fl.pidx);
 		R("FL CIDX:", fl.cidx);
-	} else if (toe_idx < toe_entries) {
-		const struct sge_ofld_rxq *rx = &adap->sge.ofldrxq[toe_idx * 4];
-		const struct sge_ofld_txq *tx = &adap->sge.ofldtxq[toe_idx * 4];
-		int n = min(4, adap->sge.ofldqsets - 4 * toe_idx);
+		RL("RxPackets:", stats.pkts);
+		RL("RxCSO:", stats.rx_cso);
+		RL("VLANxtract:", stats.vlan_ex);
+		RL("LROmerged:", stats.lro_merged);
+		RL("LROpackets:", stats.lro_pkts);
+		RL("RxDrops:", stats.rx_drops);
+		TL("TSO:", tso);
+		TL("TxCSO:", tx_cso);
+		TL("VLANins:", vlan_ins);
+		TL("TxQFull:", q.stops);
+		TL("TxQRestarts:", q.restarts);
+		TL("TxMapErr:", mapping_err);
+		RL("FLAllocErr:", fl.alloc_failed);
+		RL("FLLrgAlcErr:", fl.large_alloc_failed);
+		RL("FLStarving:", fl.starving);
+
+	} else if (iscsi_idx < iscsi_entries) {
+		const struct sge_ofld_rxq *rx =
+			&adap->sge.ofldrxq[iscsi_idx * 4];
+		const struct sge_ofld_txq *tx =
+			&adap->sge.ofldtxq[iscsi_idx * 4];
+		int n = min(4, adap->sge.ofldqsets - 4 * iscsi_idx);
 
-		S("QType:", "TOE");
+		S("QType:", "iSCSI");
 		T("TxQ ID:", q.cntxt_id);
 		T("TxQ size:", q.size);
 		T("TxQ inuse:", q.in_use);
@@ -1982,6 +2345,13 @@ do { \
 		R("FL avail:", fl.avail);
 		R("FL PIDX:", fl.pidx);
 		R("FL CIDX:", fl.cidx);
+		RL("RxPackets:", stats.pkts);
+		RL("RxImmPkts:", stats.imm);
+		RL("RxNoMem:", stats.nomem);
+		RL("FLAllocErr:", fl.alloc_failed);
+		RL("FLLrgAlcErr:", fl.large_alloc_failed);
+		RL("FLStarving:", fl.starving);
+
 	} else if (rdma_idx < rdma_entries) {
 		const struct sge_ofld_rxq *rx =
 				&adap->sge.rdmarxq[rdma_idx * 4];
@@ -2004,6 +2374,13 @@ do { \
 		R("FL avail:", fl.avail);
 		R("FL PIDX:", fl.pidx);
 		R("FL CIDX:", fl.cidx);
+		RL("RxPackets:", stats.pkts);
+		RL("RxImmPkts:", stats.imm);
+		RL("RxNoMem:", stats.nomem);
+		RL("FLAllocErr:", fl.alloc_failed);
+		RL("FLLrgAlcErr:", fl.large_alloc_failed);
+		RL("FLStarving:", fl.starving);
+
 	} else if (ciq_idx < ciq_entries) {
 		const struct sge_ofld_rxq *rx = &adap->sge.rdmaciq[ciq_idx * 4];
 		int n = min(4, adap->sge.rdmaciqs - 4 * ciq_idx);
@@ -2019,6 +2396,9 @@ do { \
 		S3("u", "Intr delay:", qtimer_val(adap, &rx[i].rspq));
 		S3("u", "Intr pktcnt:",
 		   adap->sge.counter_val[rx[i].rspq.pktcnt_idx]);
+		RL("RxAN:", stats.an);
+		RL("RxNoMem:", stats.nomem);
+
 	} else if (ctrl_idx < ctrl_entries) {
 		const struct sge_ctrl_txq *tx = &adap->sge.ctrlq[ctrl_idx * 4];
 		int n = min(4, adap->params.nports - 4 * ctrl_idx);
@@ -2029,6 +2409,8 @@ do { \
 		T("TxQ inuse:", q.in_use);
 		T("TxQ CIDX:", q.cidx);
 		T("TxQ PIDX:", q.pidx);
+		TL("TxQFull:", q.stops);
+		TL("TxQRestarts:", q.restarts);
 	} else if (fq_idx == 0) {
 		const struct sge_rspq *evtq = &adap->sge.fw_evtq;
 
@@ -2044,10 +2426,14 @@ do { \
 			   adap->sge.counter_val[evtq->pktcnt_idx]);
 	}
 #undef R
+#undef RL
 #undef T
+#undef TL
 #undef S
+#undef R3
+#undef T3
 #undef S3
-return 0;
+	return 0;
 }
 
 static int sge_queue_entries(const struct adapter *adap)
@@ -2164,6 +2550,73 @@ static const struct file_operations mem_debugfs_fops = {
 	.llseek  = default_llseek,
 };
 
+static int tid_info_show(struct seq_file *seq, void *v)
+{
+	struct adapter *adap = seq->private;
+	const struct tid_info *t = &adap->tids;
+	enum chip_type chip = CHELSIO_CHIP_VERSION(adap->params.chip);
+
+	if (t4_read_reg(adap, LE_DB_CONFIG_A) & HASHEN_F) {
+		unsigned int sb;
+
+		if (chip <= CHELSIO_T5)
+			sb = t4_read_reg(adap, LE_DB_SERVER_INDEX_A) / 4;
+		else
+			sb = t4_read_reg(adap, LE_DB_SRVR_START_INDEX_A);
+
+		if (sb) {
+			seq_printf(seq, "TID range: 0..%u/%u..%u", sb - 1,
+				   adap->tids.hash_base,
+				   t->ntids - 1);
+			seq_printf(seq, ", in use: %u/%u\n",
+				   atomic_read(&t->tids_in_use),
+				   atomic_read(&t->hash_tids_in_use));
+		} else if (adap->flags & FW_OFLD_CONN) {
+			seq_printf(seq, "TID range: %u..%u/%u..%u",
+				   t->aftid_base,
+				   t->aftid_end,
+				   adap->tids.hash_base,
+				   t->ntids - 1);
+			seq_printf(seq, ", in use: %u/%u\n",
+				   atomic_read(&t->tids_in_use),
+				   atomic_read(&t->hash_tids_in_use));
+		} else {
+			seq_printf(seq, "TID range: %u..%u",
+				   adap->tids.hash_base,
+				   t->ntids - 1);
+			seq_printf(seq, ", in use: %u\n",
+				   atomic_read(&t->hash_tids_in_use));
+		}
+	} else if (t->ntids) {
+		seq_printf(seq, "TID range: 0..%u", t->ntids - 1);
+		seq_printf(seq, ", in use: %u\n",
+			   atomic_read(&t->tids_in_use));
+	}
+
+	if (t->nstids)
+		seq_printf(seq, "STID range: %u..%u, in use: %u\n",
+			   (!t->stid_base &&
+			   (chip <= CHELSIO_T5)) ?
+			   t->stid_base + 1 : t->stid_base,
+			   t->stid_base + t->nstids - 1, t->stids_in_use);
+	if (t->natids)
+		seq_printf(seq, "ATID range: 0..%u, in use: %u\n",
+			   t->natids - 1, t->atids_in_use);
+	seq_printf(seq, "FTID range: %u..%u\n", t->ftid_base,
+		   t->ftid_base + t->nftids - 1);
+	if (t->nsftids)
+		seq_printf(seq, "SFTID range: %u..%u in use: %u\n",
+			   t->sftid_base, t->sftid_base + t->nsftids - 2,
+			   t->sftids_in_use);
+	if (t->ntids)
+		seq_printf(seq, "HW TID usage: %u IP users, %u IPv6 users\n",
+			   t4_read_reg(adap, LE_DB_ACT_CNT_IPV4_A),
+			   t4_read_reg(adap, LE_DB_ACT_CNT_IPV6_A));
+	return 0;
+}
+
+DEFINE_SIMPLE_DEBUGFS_FILE(tid_info);
+
 static void add_debugfs_mem(struct adapter *adap, const char *name,
 			    unsigned int idx, unsigned int size_mb)
 {
@@ -2227,6 +2680,290 @@ static const struct file_operations blocked_fl_fops = {
 	.llseek  = generic_file_llseek,
 };
 
+struct mem_desc {
+	unsigned int base;
+	unsigned int limit;
+	unsigned int idx;
+};
+
+static int mem_desc_cmp(const void *a, const void *b)
+{
+	return ((const struct mem_desc *)a)->base -
+	       ((const struct mem_desc *)b)->base;
+}
+
+static void mem_region_show(struct seq_file *seq, const char *name,
+			    unsigned int from, unsigned int to)
+{
+	char buf[40];
+
+	string_get_size((u64)to - from + 1, 1, STRING_UNITS_2, buf,
+			sizeof(buf));
+	seq_printf(seq, "%-15s %#x-%#x [%s]\n", name, from, to, buf);
+}
+
+static int meminfo_show(struct seq_file *seq, void *v)
+{
+	static const char * const memory[] = { "EDC0:", "EDC1:", "MC:",
+					"MC0:", "MC1:"};
+	static const char * const region[] = {
+		"DBQ contexts:", "IMSG contexts:", "FLM cache:", "TCBs:",
+		"Pstructs:", "Timers:", "Rx FL:", "Tx FL:", "Pstruct FL:",
+		"Tx payload:", "Rx payload:", "LE hash:", "iSCSI region:",
+		"TDDP region:", "TPT region:", "STAG region:", "RQ region:",
+		"RQUDP region:", "PBL region:", "TXPBL region:",
+		"DBVFIFO region:", "ULPRX state:", "ULPTX state:",
+		"On-chip queues:"
+	};
+
+	int i, n;
+	u32 lo, hi, used, alloc;
+	struct mem_desc avail[4];
+	struct mem_desc mem[ARRAY_SIZE(region) + 3];      /* up to 3 holes */
+	struct mem_desc *md = mem;
+	struct adapter *adap = seq->private;
+
+	for (i = 0; i < ARRAY_SIZE(mem); i++) {
+		mem[i].limit = 0;
+		mem[i].idx = i;
+	}
+
+	/* Find and sort the populated memory ranges */
+	i = 0;
+	lo = t4_read_reg(adap, MA_TARGET_MEM_ENABLE_A);
+	if (lo & EDRAM0_ENABLE_F) {
+		hi = t4_read_reg(adap, MA_EDRAM0_BAR_A);
+		avail[i].base = EDRAM0_BASE_G(hi) << 20;
+		avail[i].limit = avail[i].base + (EDRAM0_SIZE_G(hi) << 20);
+		avail[i].idx = 0;
+		i++;
+	}
+	if (lo & EDRAM1_ENABLE_F) {
+		hi = t4_read_reg(adap, MA_EDRAM1_BAR_A);
+		avail[i].base = EDRAM1_BASE_G(hi) << 20;
+		avail[i].limit = avail[i].base + (EDRAM1_SIZE_G(hi) << 20);
+		avail[i].idx = 1;
+		i++;
+	}
+
+	if (is_t5(adap->params.chip)) {
+		if (lo & EXT_MEM0_ENABLE_F) {
+			hi = t4_read_reg(adap, MA_EXT_MEMORY0_BAR_A);
+			avail[i].base = EXT_MEM0_BASE_G(hi) << 20;
+			avail[i].limit =
+				avail[i].base + (EXT_MEM0_SIZE_G(hi) << 20);
+			avail[i].idx = 3;
+			i++;
+		}
+		if (lo & EXT_MEM1_ENABLE_F) {
+			hi = t4_read_reg(adap, MA_EXT_MEMORY1_BAR_A);
+			avail[i].base = EXT_MEM1_BASE_G(hi) << 20;
+			avail[i].limit =
+				avail[i].base + (EXT_MEM1_SIZE_G(hi) << 20);
+			avail[i].idx = 4;
+			i++;
+		}
+	} else {
+		if (lo & EXT_MEM_ENABLE_F) {
+			hi = t4_read_reg(adap, MA_EXT_MEMORY_BAR_A);
+			avail[i].base = EXT_MEM_BASE_G(hi) << 20;
+			avail[i].limit =
+				avail[i].base + (EXT_MEM_SIZE_G(hi) << 20);
+			avail[i].idx = 2;
+			i++;
+		}
+	}
+	if (!i)                                    /* no memory available */
+		return 0;
+	sort(avail, i, sizeof(struct mem_desc), mem_desc_cmp, NULL);
+
+	(md++)->base = t4_read_reg(adap, SGE_DBQ_CTXT_BADDR_A);
+	(md++)->base = t4_read_reg(adap, SGE_IMSG_CTXT_BADDR_A);
+	(md++)->base = t4_read_reg(adap, SGE_FLM_CACHE_BADDR_A);
+	(md++)->base = t4_read_reg(adap, TP_CMM_TCB_BASE_A);
+	(md++)->base = t4_read_reg(adap, TP_CMM_MM_BASE_A);
+	(md++)->base = t4_read_reg(adap, TP_CMM_TIMER_BASE_A);
+	(md++)->base = t4_read_reg(adap, TP_CMM_MM_RX_FLST_BASE_A);
+	(md++)->base = t4_read_reg(adap, TP_CMM_MM_TX_FLST_BASE_A);
+	(md++)->base = t4_read_reg(adap, TP_CMM_MM_PS_FLST_BASE_A);
+
+	/* the next few have explicit upper bounds */
+	md->base = t4_read_reg(adap, TP_PMM_TX_BASE_A);
+	md->limit = md->base - 1 +
+		    t4_read_reg(adap, TP_PMM_TX_PAGE_SIZE_A) *
+		    PMTXMAXPAGE_G(t4_read_reg(adap, TP_PMM_TX_MAX_PAGE_A));
+	md++;
+
+	md->base = t4_read_reg(adap, TP_PMM_RX_BASE_A);
+	md->limit = md->base - 1 +
+		    t4_read_reg(adap, TP_PMM_RX_PAGE_SIZE_A) *
+		    PMRXMAXPAGE_G(t4_read_reg(adap, TP_PMM_RX_MAX_PAGE_A));
+	md++;
+
+	if (t4_read_reg(adap, LE_DB_CONFIG_A) & HASHEN_F) {
+		if (CHELSIO_CHIP_VERSION(adap->params.chip) <= CHELSIO_T5) {
+			hi = t4_read_reg(adap, LE_DB_TID_HASHBASE_A) / 4;
+			md->base = t4_read_reg(adap, LE_DB_HASH_TID_BASE_A);
+		 } else {
+			hi = t4_read_reg(adap, LE_DB_HASH_TID_BASE_A);
+			md->base = t4_read_reg(adap,
+					       LE_DB_HASH_TBL_BASE_ADDR_A);
+		}
+		md->limit = 0;
+	} else {
+		md->base = 0;
+		md->idx = ARRAY_SIZE(region);  /* hide it */
+	}
+	md++;
+
+#define ulp_region(reg) do { \
+	md->base = t4_read_reg(adap, ULP_ ## reg ## _LLIMIT_A);\
+	(md++)->limit = t4_read_reg(adap, ULP_ ## reg ## _ULIMIT_A); \
+} while (0)
+
+	ulp_region(RX_ISCSI);
+	ulp_region(RX_TDDP);
+	ulp_region(TX_TPT);
+	ulp_region(RX_STAG);
+	ulp_region(RX_RQ);
+	ulp_region(RX_RQUDP);
+	ulp_region(RX_PBL);
+	ulp_region(TX_PBL);
+#undef ulp_region
+	md->base = 0;
+	md->idx = ARRAY_SIZE(region);
+	if (!is_t4(adap->params.chip)) {
+		u32 size = 0;
+		u32 sge_ctrl = t4_read_reg(adap, SGE_CONTROL2_A);
+		u32 fifo_size = t4_read_reg(adap, SGE_DBVFIFO_SIZE_A);
+
+		if (is_t5(adap->params.chip)) {
+			if (sge_ctrl & VFIFO_ENABLE_F)
+				size = DBVFIFO_SIZE_G(fifo_size);
+		} else {
+			size = T6_DBVFIFO_SIZE_G(fifo_size);
+		}
+
+		if (size) {
+			md->base = BASEADDR_G(t4_read_reg(adap,
+					SGE_DBVFIFO_BADDR_A));
+			md->limit = md->base + (size << 2) - 1;
+		}
+	}
+
+	md++;
+
+	md->base = t4_read_reg(adap, ULP_RX_CTX_BASE_A);
+	md->limit = 0;
+	md++;
+	md->base = t4_read_reg(adap, ULP_TX_ERR_TABLE_BASE_A);
+	md->limit = 0;
+	md++;
+
+	md->base = adap->vres.ocq.start;
+	if (adap->vres.ocq.size)
+		md->limit = md->base + adap->vres.ocq.size - 1;
+	else
+		md->idx = ARRAY_SIZE(region);  /* hide it */
+	md++;
+
+	/* add any address-space holes, there can be up to 3 */
+	for (n = 0; n < i - 1; n++)
+		if (avail[n].limit < avail[n + 1].base)
+			(md++)->base = avail[n].limit;
+	if (avail[n].limit)
+		(md++)->base = avail[n].limit;
+
+	n = md - mem;
+	sort(mem, n, sizeof(struct mem_desc), mem_desc_cmp, NULL);
+
+	for (lo = 0; lo < i; lo++)
+		mem_region_show(seq, memory[avail[lo].idx], avail[lo].base,
+				avail[lo].limit - 1);
+
+	seq_putc(seq, '\n');
+	for (i = 0; i < n; i++) {
+		if (mem[i].idx >= ARRAY_SIZE(region))
+			continue;                        /* skip holes */
+		if (!mem[i].limit)
+			mem[i].limit = i < n - 1 ? mem[i + 1].base - 1 : ~0;
+		mem_region_show(seq, region[mem[i].idx], mem[i].base,
+				mem[i].limit);
+	}
+
+	seq_putc(seq, '\n');
+	lo = t4_read_reg(adap, CIM_SDRAM_BASE_ADDR_A);
+	hi = t4_read_reg(adap, CIM_SDRAM_ADDR_SIZE_A) + lo - 1;
+	mem_region_show(seq, "uP RAM:", lo, hi);
+
+	lo = t4_read_reg(adap, CIM_EXTMEM2_BASE_ADDR_A);
+	hi = t4_read_reg(adap, CIM_EXTMEM2_ADDR_SIZE_A) + lo - 1;
+	mem_region_show(seq, "uP Extmem2:", lo, hi);
+
+	lo = t4_read_reg(adap, TP_PMM_RX_MAX_PAGE_A);
+	seq_printf(seq, "\n%u Rx pages of size %uKiB for %u channels\n",
+		   PMRXMAXPAGE_G(lo),
+		   t4_read_reg(adap, TP_PMM_RX_PAGE_SIZE_A) >> 10,
+		   (lo & PMRXNUMCHN_F) ? 2 : 1);
+
+	lo = t4_read_reg(adap, TP_PMM_TX_MAX_PAGE_A);
+	hi = t4_read_reg(adap, TP_PMM_TX_PAGE_SIZE_A);
+	seq_printf(seq, "%u Tx pages of size %u%ciB for %u channels\n",
+		   PMTXMAXPAGE_G(lo),
+		   hi >= (1 << 20) ? (hi >> 20) : (hi >> 10),
+		   hi >= (1 << 20) ? 'M' : 'K', 1 << PMTXNUMCHN_G(lo));
+	seq_printf(seq, "%u p-structs\n\n",
+		   t4_read_reg(adap, TP_CMM_MM_MAX_PSTRUCT_A));
+
+	for (i = 0; i < 4; i++) {
+		if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5)
+			lo = t4_read_reg(adap, MPS_RX_MAC_BG_PG_CNT0_A + i * 4);
+		else
+			lo = t4_read_reg(adap, MPS_RX_PG_RSV0_A + i * 4);
+		if (is_t5(adap->params.chip)) {
+			used = T5_USED_G(lo);
+			alloc = T5_ALLOC_G(lo);
+		} else {
+			used = USED_G(lo);
+			alloc = ALLOC_G(lo);
+		}
+		/* For T6 these are MAC buffer groups */
+		seq_printf(seq, "Port %d using %u pages out of %u allocated\n",
+			   i, used, alloc);
+	}
+	for (i = 0; i < adap->params.arch.nchan; i++) {
+		if (CHELSIO_CHIP_VERSION(adap->params.chip) > CHELSIO_T5)
+			lo = t4_read_reg(adap,
+					 MPS_RX_LPBK_BG_PG_CNT0_A + i * 4);
+		else
+			lo = t4_read_reg(adap, MPS_RX_PG_RSV4_A + i * 4);
+		if (is_t5(adap->params.chip)) {
+			used = T5_USED_G(lo);
+			alloc = T5_ALLOC_G(lo);
+		} else {
+			used = USED_G(lo);
+			alloc = ALLOC_G(lo);
+		}
+		/* For T6 these are MAC buffer groups */
+		seq_printf(seq,
+			   "Loopback %d using %u pages out of %u allocated\n",
+			   i, used, alloc);
+	}
+	return 0;
+}
+
+static int meminfo_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, meminfo_show, inode->i_private);
+}
+
+static const struct file_operations meminfo_fops = {
+	.owner   = THIS_MODULE,
+	.open    = meminfo_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = single_release,
+};
 /* Add an array of Debug FS files.
  */
 void add_debugfs_files(struct adapter *adap,
@@ -2264,6 +3001,10 @@ int t4_setup_debugfs(struct adapter *adap)
 		{ "mbox5", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 5 },
 		{ "mbox6", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 6 },
 		{ "mbox7", &mbox_debugfs_fops, S_IRUSR | S_IWUSR, 7 },
+		{ "trace0", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 0 },
+		{ "trace1", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 1 },
+		{ "trace2", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 2 },
+		{ "trace3", &mps_trc_debugfs_fops, S_IRUSR | S_IWUSR, 3 },
 		{ "l2t", &t4_l2t_fops, S_IRUSR, 0},
 		{ "mps_tcam", &mps_tcam_debugfs_fops, S_IRUSR, 0 },
 		{ "rss", &rss_debugfs_fops, S_IRUSR, 0 },
@@ -2293,7 +3034,9 @@ int t4_setup_debugfs(struct adapter *adap)
 #if IS_ENABLED(CONFIG_IPV6)
 		{ "clip_tbl", &clip_tbl_debugfs_fops, S_IRUSR, 0 },
 #endif
+		{ "tids", &tid_info_debugfs_fops, S_IRUSR, 0},
 		{ "blocked_fl", &blocked_fl_fops, S_IRUSR | S_IWUSR, 0 },
+		{ "meminfo", &meminfo_fops, S_IRUSR, 0 },
 	};
 
 	/* Debug FS nodes common to all T5 and later adapters.
@@ -2341,6 +3084,10 @@ int t4_setup_debugfs(struct adapter *adap)
 
 	de = debugfs_create_file_size("flash", S_IRUSR, adap->debugfs_root, adap,
 				      &flash_debugfs_fops, adap->params.sf_size);
+	debugfs_create_bool("use_backdoor", S_IWUSR | S_IRUSR,
+			    adap->debugfs_root, &adap->use_bd);
+	debugfs_create_bool("trace_rss", S_IWUSR | S_IRUSR,
+			    adap->debugfs_root, &adap->trace_rss);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
index 687acf71fa15..5eedb98ff581 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
@@ -925,6 +925,20 @@ static int set_flash(struct net_device *netdev, struct ethtool_flash *ef)
 	const struct firmware *fw;
 	struct adapter *adap = netdev2adap(netdev);
 	unsigned int mbox = PCIE_FW_MASTER_M + 1;
+	u32 pcie_fw;
+	unsigned int master;
+	u8 master_vld = 0;
+
+	pcie_fw = t4_read_reg(adap, PCIE_FW_A);
+	master = PCIE_FW_MASTER_G(pcie_fw);
+	if (pcie_fw & PCIE_FW_MASTER_VLD_F)
+		master_vld = 1;
+	/* if csiostor is the master return */
+	if (master_vld && (master != adap->pf)) {
+		dev_warn(adap->pdev_dev,
+			 "cxgb4 driver needs to be loaded as MASTER to support FW flash\n");
+		return -EOPNOTSUPP;
+	}
 
 	ef->data[sizeof(ef->data) - 1] = '\0';
 	ret = request_firmware(&fw, ef->data, adap->pdev_dev);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index 351f3b1bf800..eb22d58743e2 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -1548,7 +1548,7 @@ int cxgb4_alloc_sftid(struct tid_info *t, int family, void *data)
 		t->stid_tab[stid].data = data;
 		stid -= t->nstids;
 		stid += t->sftid_base;
-		t->stids_in_use++;
+		t->sftids_in_use++;
 	}
 	spin_unlock_bh(&t->stid_lock);
 	return stid;
@@ -1573,10 +1573,14 @@ void cxgb4_free_stid(struct tid_info *t, unsigned int stid, int family)
 	else
 		bitmap_release_region(t->stid_bmap, stid, 2);
 	t->stid_tab[stid].data = NULL;
-	if (family == PF_INET)
-		t->stids_in_use--;
-	else
-		t->stids_in_use -= 4;
+	if (stid < t->nstids) {
+		if (family == PF_INET)
+			t->stids_in_use--;
+		else
+			t->stids_in_use -= 4;
+	} else {
+		t->sftids_in_use--;
+	}
 	spin_unlock_bh(&t->stid_lock);
 }
 EXPORT_SYMBOL(cxgb4_free_stid);
@@ -1654,20 +1658,25 @@ static void process_tid_release_list(struct work_struct *work)
  */
 void cxgb4_remove_tid(struct tid_info *t, unsigned int chan, unsigned int tid)
 {
-	void *old;
 	struct sk_buff *skb;
 	struct adapter *adap = container_of(t, struct adapter, tids);
 
-	old = t->tid_tab[tid];
+	WARN_ON(tid >= t->ntids);
+
+	if (t->tid_tab[tid]) {
+		t->tid_tab[tid] = NULL;
+		if (t->hash_base && (tid >= t->hash_base))
+			atomic_dec(&t->hash_tids_in_use);
+		else
+			atomic_dec(&t->tids_in_use);
+	}
+
 	skb = alloc_skb(sizeof(struct cpl_tid_release), GFP_ATOMIC);
 	if (likely(skb)) {
-		t->tid_tab[tid] = NULL;
 		mk_tid_release(skb, chan, tid);
 		t4_ofld_send(adap, skb);
 	} else
 		cxgb4_queue_tid_release(t, chan, tid);
-	if (old)
-		atomic_dec(&t->tids_in_use);
 }
 EXPORT_SYMBOL(cxgb4_remove_tid);
 
@@ -1702,9 +1711,11 @@ static int tid_init(struct tid_info *t)
 	spin_lock_init(&t->atid_lock);
 
 	t->stids_in_use = 0;
+	t->sftids_in_use = 0;
 	t->afree = NULL;
 	t->atids_in_use = 0;
 	atomic_set(&t->tids_in_use, 0);
+	atomic_set(&t->hash_tids_in_use, 0);
 
 	/* Setup the free list for atid_tab and clear the stid bitmap. */
 	if (natids) {
@@ -3657,6 +3668,10 @@ static int adap_init0(struct adapter *adap)
 	 */
 	t4_get_fw_version(adap, &adap->params.fw_vers);
 	t4_get_tp_version(adap, &adap->params.tp_vers);
+	ret = t4_check_fw_version(adap);
+	/* If firmware is too old (not supported by driver) force an update. */
+	if (ret == -EFAULT)
+		state = DEV_STATE_UNINIT;
 	if ((adap->flags & MASTER_PF) && state != DEV_STATE_INIT) {
 		struct fw_info *fw_info;
 		struct fw_hdr *card_fw;
@@ -4551,6 +4566,32 @@ static void free_some_resources(struct adapter *adapter)
 		   NETIF_F_IPV6_CSUM | NETIF_F_HIGHDMA)
 #define SEGMENT_SIZE 128
 
+static int get_chip_type(struct pci_dev *pdev, u32 pl_rev)
+{
+	int ver, chip;
+	u16 device_id;
+
+	/* Retrieve adapter's device ID */
+	pci_read_config_word(pdev, PCI_DEVICE_ID, &device_id);
+	ver = device_id >> 12;
+	switch (ver) {
+	case CHELSIO_T4:
+		chip |= CHELSIO_CHIP_CODE(CHELSIO_T4, pl_rev);
+		break;
+	case CHELSIO_T5:
+		chip |= CHELSIO_CHIP_CODE(CHELSIO_T5, pl_rev);
+		break;
+	case CHELSIO_T6:
+		chip |= CHELSIO_CHIP_CODE(CHELSIO_T6, pl_rev);
+		break;
+	default:
+		dev_err(&pdev->dev, "Device %d is not supported\n",
+			device_id);
+		return -EINVAL;
+	}
+	return chip;
+}
+
 static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	int func, i, err, s_qpp, qpp, num_seg;
@@ -4558,6 +4599,8 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	bool highdma = false;
 	struct adapter *adapter = NULL;
 	void __iomem *regs;
+	u32 whoami, pl_rev;
+	enum chip_type chip;
 
 	printk_once(KERN_INFO "%s - version %s\n", DRV_DESC, DRV_VERSION);
 
@@ -4586,7 +4629,11 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto out_unmap_bar0;
 
 	/* We control everything through one PF */
-	func = SOURCEPF_G(readl(regs + PL_WHOAMI_A));
+	whoami = readl(regs + PL_WHOAMI_A);
+	pl_rev = REV_G(readl(regs + PL_REV_A));
+	chip = get_chip_type(pdev, pl_rev);
+	func = CHELSIO_CHIP_VERSION(chip) <= CHELSIO_T5 ?
+		SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami);
 	if (func != ent->driver_data) {
 		iounmap(regs);
 		pci_disable_device(pdev);
@@ -4757,7 +4804,7 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 */
 	cfg_queues(adapter);
 
-	adapter->l2t = t4_init_l2t();
+	adapter->l2t = t4_init_l2t(adapter->l2t_start, adapter->l2t_end);
 	if (!adapter->l2t) {
 		/* We tolerate a lack of L2T, giving up some functionality */
 		dev_warn(&pdev->dev, "could not allocate L2T, continuing\n");
@@ -4782,6 +4829,22 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		adapter->params.offload = 0;
 	}
 
+	if (is_offload(adapter)) {
+		if (t4_read_reg(adapter, LE_DB_CONFIG_A) & HASHEN_F) {
+			u32 hash_base, hash_reg;
+
+			if (chip <= CHELSIO_T5) {
+				hash_reg = LE_DB_TID_HASHBASE_A;
+				hash_base = t4_read_reg(adapter, hash_reg);
+				adapter->tids.hash_base = hash_base / 4;
+			} else {
+				hash_reg = T6_LE_DB_HASH_TID_BASE_A;
+				hash_base = t4_read_reg(adapter, hash_reg);
+				adapter->tids.hash_base = hash_base;
+			}
+		}
+	}
+
 	/* See what interrupts we'll be using */
 	if (msi > 1 && enable_msix(adapter) == 0)
 		adapter->flags |= USING_MSIX;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index b27897d4f787..c3a8be5541e7 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -96,6 +96,7 @@ struct tid_info {
 	unsigned long *stid_bmap;
 	unsigned int nstids;
 	unsigned int stid_base;
+	unsigned int hash_base;
 
 	union aopen_entry *atid_tab;
 	unsigned int natids;
@@ -116,8 +117,12 @@ struct tid_info {
 
 	spinlock_t stid_lock;
 	unsigned int stids_in_use;
+	unsigned int sftids_in_use;
 
+	/* TIDs in the TCAM */
 	atomic_t tids_in_use;
+	/* TIDs in the HASH */
+	atomic_t hash_tids_in_use;
 };
 
 static inline void *lookup_tid(const struct tid_info *t, unsigned int tid)
@@ -147,7 +152,10 @@ static inline void cxgb4_insert_tid(struct tid_info *t, void *data,
 				    unsigned int tid)
 {
 	t->tid_tab[tid] = data;
-	atomic_inc(&t->tids_in_use);
+	if (t->hash_base && (tid >= t->hash_base))
+		atomic_inc(&t->hash_tids_in_use);
+	else
+		atomic_inc(&t->tids_in_use);
 }
 
 int cxgb4_alloc_atid(struct tid_info *t, void *data);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/l2t.c b/drivers/net/ethernet/chelsio/cxgb4/l2t.c
index 252efc29321f..ac27898c6ab0 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/l2t.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/l2t.c
@@ -51,24 +51,17 @@
 #define VLAN_NONE 0xfff
 
 /* identifies sync vs async L2T_WRITE_REQs */
-#define F_SYNC_WR    (1 << 12)
-
-enum {
-	L2T_STATE_VALID,      /* entry is up to date */
-	L2T_STATE_STALE,      /* entry may be used but needs revalidation */
-	L2T_STATE_RESOLVING,  /* entry needs address resolution */
-	L2T_STATE_SYNC_WRITE, /* synchronous write of entry underway */
-
-	/* when state is one of the below the entry is not hashed */
-	L2T_STATE_SWITCHING,  /* entry is being used by a switching filter */
-	L2T_STATE_UNUSED      /* entry not in use */
-};
+#define SYNC_WR_S    12
+#define SYNC_WR_V(x) ((x) << SYNC_WR_S)
+#define SYNC_WR_F    SYNC_WR_V(1)
 
 struct l2t_data {
+	unsigned int l2t_start;     /* start index of our piece of the L2T */
+	unsigned int l2t_size;      /* number of entries in l2tab */
 	rwlock_t lock;
 	atomic_t nfree;             /* number of free entries */
 	struct l2t_entry *rover;    /* starting point for next allocation */
-	struct l2t_entry l2tab[L2T_SIZE];
+	struct l2t_entry l2tab[0];  /* MUST BE LAST */
 };
 
 static inline unsigned int vlan_prio(const struct l2t_entry *e)
@@ -85,29 +78,36 @@ static inline void l2t_hold(struct l2t_data *d, struct l2t_entry *e)
 /*
  * To avoid having to check address families we do not allow v4 and v6
  * neighbors to be on the same hash chain.  We keep v4 entries in the first
- * half of available hash buckets and v6 in the second.
+ * half of available hash buckets and v6 in the second.  We need at least two
+ * entries in our L2T for this scheme to work.
  */
 enum {
-	L2T_SZ_HALF = L2T_SIZE / 2,
-	L2T_HASH_MASK = L2T_SZ_HALF - 1
+	L2T_MIN_HASH_BUCKETS = 2,
 };
 
-static inline unsigned int arp_hash(const u32 *key, int ifindex)
+static inline unsigned int arp_hash(struct l2t_data *d, const u32 *key,
+				    int ifindex)
 {
-	return jhash_2words(*key, ifindex, 0) & L2T_HASH_MASK;
+	unsigned int l2t_size_half = d->l2t_size / 2;
+
+	return jhash_2words(*key, ifindex, 0) % l2t_size_half;
 }
 
-static inline unsigned int ipv6_hash(const u32 *key, int ifindex)
+static inline unsigned int ipv6_hash(struct l2t_data *d, const u32 *key,
+				     int ifindex)
 {
+	unsigned int l2t_size_half = d->l2t_size / 2;
 	u32 xor = key[0] ^ key[1] ^ key[2] ^ key[3];
 
-	return L2T_SZ_HALF + (jhash_2words(xor, ifindex, 0) & L2T_HASH_MASK);
+	return (l2t_size_half +
+		(jhash_2words(xor, ifindex, 0) % l2t_size_half));
 }
 
-static unsigned int addr_hash(const u32 *addr, int addr_len, int ifindex)
+static unsigned int addr_hash(struct l2t_data *d, const u32 *addr,
+			      int addr_len, int ifindex)
 {
-	return addr_len == 4 ? arp_hash(addr, ifindex) :
-			       ipv6_hash(addr, ifindex);
+	return addr_len == 4 ? arp_hash(d, addr, ifindex) :
+			       ipv6_hash(d, addr, ifindex);
 }
 
 /*
@@ -139,6 +139,8 @@ static void neigh_replace(struct l2t_entry *e, struct neighbour *n)
  */
 static int write_l2e(struct adapter *adap, struct l2t_entry *e, int sync)
 {
+	struct l2t_data *d = adap->l2t;
+	unsigned int l2t_idx = e->idx + d->l2t_start;
 	struct sk_buff *skb;
 	struct cpl_l2t_write_req *req;
 
@@ -150,10 +152,10 @@ static int write_l2e(struct adapter *adap, struct l2t_entry *e, int sync)
 	INIT_TP_WR(req, 0);
 
 	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_L2T_WRITE_REQ,
-					e->idx | (sync ? F_SYNC_WR : 0) |
+					l2t_idx | (sync ? SYNC_WR_F : 0) |
 					TID_QID_V(adap->sge.fw_evtq.abs_id)));
 	req->params = htons(L2T_W_PORT_V(e->lport) | L2T_W_NOREPLY_V(!sync));
-	req->l2t_idx = htons(e->idx);
+	req->l2t_idx = htons(l2t_idx);
 	req->vlan = htons(e->vlan);
 	if (e->neigh && !(e->neigh->dev->flags & IFF_LOOPBACK))
 		memcpy(e->dmac, e->neigh->ha, sizeof(e->dmac));
@@ -190,18 +192,19 @@ static void send_pending(struct adapter *adap, struct l2t_entry *e)
  */
 void do_l2t_write_rpl(struct adapter *adap, const struct cpl_l2t_write_rpl *rpl)
 {
+	struct l2t_data *d = adap->l2t;
 	unsigned int tid = GET_TID(rpl);
-	unsigned int idx = tid & (L2T_SIZE - 1);
+	unsigned int l2t_idx = tid % L2T_SIZE;
 
 	if (unlikely(rpl->status != CPL_ERR_NONE)) {
 		dev_err(adap->pdev_dev,
 			"Unexpected L2T_WRITE_RPL status %u for entry %u\n",
-			rpl->status, idx);
+			rpl->status, l2t_idx);
 		return;
 	}
 
-	if (tid & F_SYNC_WR) {
-		struct l2t_entry *e = &adap->l2t->l2tab[idx];
+	if (tid & SYNC_WR_F) {
+		struct l2t_entry *e = &d->l2tab[l2t_idx - d->l2t_start];
 
 		spin_lock(&e->lock);
 		if (e->state != L2T_STATE_SWITCHING) {
@@ -276,7 +279,7 @@ static struct l2t_entry *alloc_l2e(struct l2t_data *d)
 		return NULL;
 
 	/* there's definitely a free entry */
-	for (e = d->rover, end = &d->l2tab[L2T_SIZE]; e != end; ++e)
+	for (e = d->rover, end = &d->l2tab[d->l2t_size]; e != end; ++e)
 		if (atomic_read(&e->refcnt) == 0)
 			goto found;
 
@@ -368,7 +371,7 @@ struct l2t_entry *cxgb4_l2t_get(struct l2t_data *d, struct neighbour *neigh,
 	int addr_len = neigh->tbl->key_len;
 	u32 *addr = (u32 *)neigh->primary_key;
 	int ifidx = neigh->dev->ifindex;
-	int hash = addr_hash(addr, addr_len, ifidx);
+	int hash = addr_hash(d, addr, addr_len, ifidx);
 
 	if (neigh->dev->flags & IFF_LOOPBACK)
 		lport = netdev2pinfo(physdev)->tx_chan + 4;
@@ -481,7 +484,7 @@ void t4_l2t_update(struct adapter *adap, struct neighbour *neigh)
 	int addr_len = neigh->tbl->key_len;
 	u32 *addr = (u32 *) neigh->primary_key;
 	int ifidx = neigh->dev->ifindex;
-	int hash = addr_hash(addr, addr_len, ifidx);
+	int hash = addr_hash(d, addr, addr_len, ifidx);
 
 	read_lock_bh(&d->lock);
 	for (e = d->l2tab[hash].first; e; e = e->next)
@@ -554,20 +557,30 @@ int t4_l2t_set_switching(struct adapter *adap, struct l2t_entry *e, u16 vlan,
 	return write_l2e(adap, e, 0);
 }
 
-struct l2t_data *t4_init_l2t(void)
+struct l2t_data *t4_init_l2t(unsigned int l2t_start, unsigned int l2t_end)
 {
+	unsigned int l2t_size;
 	int i;
 	struct l2t_data *d;
 
-	d = t4_alloc_mem(sizeof(*d));
+	if (l2t_start >= l2t_end || l2t_end >= L2T_SIZE)
+		return NULL;
+	l2t_size = l2t_end - l2t_start + 1;
+	if (l2t_size < L2T_MIN_HASH_BUCKETS)
+		return NULL;
+
+	d = t4_alloc_mem(sizeof(*d) + l2t_size * sizeof(struct l2t_entry));
 	if (!d)
 		return NULL;
 
+	d->l2t_start = l2t_start;
+	d->l2t_size = l2t_size;
+
 	d->rover = d->l2tab;
-	atomic_set(&d->nfree, L2T_SIZE);
+	atomic_set(&d->nfree, l2t_size);
 	rwlock_init(&d->lock);
 
-	for (i = 0; i < L2T_SIZE; ++i) {
+	for (i = 0; i < d->l2t_size; ++i) {
 		d->l2tab[i].idx = i;
 		d->l2tab[i].state = L2T_STATE_UNUSED;
 		spin_lock_init(&d->l2tab[i].lock);
@@ -578,9 +591,9 @@ struct l2t_data *t4_init_l2t(void)
 
 static inline void *l2t_get_idx(struct seq_file *seq, loff_t pos)
 {
-	struct l2t_entry *l2tab = seq->private;
+	struct l2t_data *d = seq->private;
 
-	return pos >= L2T_SIZE ? NULL : &l2tab[pos];
+	return pos >= d->l2t_size ? NULL : &d->l2tab[pos];
 }
 
 static void *l2t_seq_start(struct seq_file *seq, loff_t *pos)
@@ -620,6 +633,7 @@ static int l2t_seq_show(struct seq_file *seq, void *v)
 			 "Ethernet address  VLAN/P LP State Users Port\n");
 	else {
 		char ip[60];
+		struct l2t_data *d = seq->private;
 		struct l2t_entry *e = v;
 
 		spin_lock_bh(&e->lock);
@@ -628,7 +642,7 @@ static int l2t_seq_show(struct seq_file *seq, void *v)
 		else
 			sprintf(ip, e->v6 ? "%pI6c" : "%pI4", e->addr);
 		seq_printf(seq, "%4u %-25s %17pM %4d %u %2u   %c   %5u %s\n",
-			   e->idx, ip, e->dmac,
+			   e->idx + d->l2t_start, ip, e->dmac,
 			   e->vlan & VLAN_VID_MASK, vlan_prio(e), e->lport,
 			   l2e_state(e), atomic_read(&e->refcnt),
 			   e->neigh ? e->neigh->dev->name : "");
@@ -652,7 +666,7 @@ static int l2t_seq_open(struct inode *inode, struct file *file)
 		struct adapter *adap = inode->i_private;
 		struct seq_file *seq = file->private_data;
 
-		seq->private = adap->l2t->l2tab;
+		seq->private = adap->l2t;
 	}
 	return rc;
 }
diff --git a/drivers/net/ethernet/chelsio/cxgb4/l2t.h b/drivers/net/ethernet/chelsio/cxgb4/l2t.h
index a30126ce90cb..b38dc526aad5 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/l2t.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/l2t.h
@@ -39,6 +39,20 @@
 #include <linux/if_ether.h>
 #include <linux/atomic.h>
 
+enum { L2T_SIZE = 4096 };     /* # of L2T entries */
+
+enum {
+	L2T_STATE_VALID,      /* entry is up to date */
+	L2T_STATE_STALE,      /* entry may be used but needs revalidation */
+	L2T_STATE_RESOLVING,  /* entry needs address resolution */
+	L2T_STATE_SYNC_WRITE, /* synchronous write of entry underway */
+	L2T_STATE_NOARP,      /* Netdev down or removed*/
+
+	/* when state is one of the below the entry is not hashed */
+	L2T_STATE_SWITCHING,  /* entry is being used by a switching filter */
+	L2T_STATE_UNUSED      /* entry not in use */
+};
+
 struct adapter;
 struct l2t_data;
 struct neighbour;
@@ -56,7 +70,7 @@ struct cpl_l2t_write_rpl;
  */
 struct l2t_entry {
 	u16 state;                  /* entry state */
-	u16 idx;                    /* entry index */
+	u16 idx;                    /* entry index within in-memory table */
 	u32 addr[4];                /* next hop IP or IPv6 address */
 	int ifindex;                /* neighbor's net_device's ifindex */
 	struct neighbour *neigh;    /* associated neighbour */
@@ -104,7 +118,7 @@ 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);
+struct l2t_data *t4_init_l2t(unsigned int l2t_start, unsigned int l2t_end);
 void do_l2t_write_rpl(struct adapter *p, const struct cpl_l2t_write_rpl *rpl);
 
 extern const struct file_operations t4_l2t_fops;
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index 942db078f33a..78f446c58422 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -1137,7 +1137,7 @@ cxgb_fcoe_offload(struct sk_buff *skb, struct adapter *adap,
  */
 netdev_tx_t t4_eth_xmit(struct sk_buff *skb, struct net_device *dev)
 {
-	u32 wr_mid;
+	u32 wr_mid, ctrl0;
 	u64 cntrl, *end;
 	int qidx, credits;
 	unsigned int flits, ndesc;
@@ -1274,9 +1274,15 @@ out_free:	dev_kfree_skb_any(skb);
 #endif /* CONFIG_CHELSIO_T4_FCOE */
 	}
 
-	cpl->ctrl0 = htonl(TXPKT_OPCODE_V(CPL_TX_PKT_XT) |
-			   TXPKT_INTF_V(pi->tx_chan) |
-			   TXPKT_PF_V(adap->pf));
+	ctrl0 = TXPKT_OPCODE_V(CPL_TX_PKT_XT) | TXPKT_INTF_V(pi->tx_chan) |
+		TXPKT_PF_V(adap->pf);
+#ifdef CONFIG_CHELSIO_T4_DCB
+	if (is_t4(adap->params.chip))
+		ctrl0 |= TXPKT_OVLAN_IDX_V(q->dcb_prio);
+	else
+		ctrl0 |= TXPKT_T5_OVLAN_IDX_V(q->dcb_prio);
+#endif
+	cpl->ctrl0 = htonl(ctrl0);
 	cpl->pack = htons(0);
 	cpl->len = htons(skb->len);
 	cpl->ctrl1 = cpu_to_be64(cntrl);
@@ -1418,18 +1424,17 @@ static void restart_ctrlq(unsigned long data)
 		struct fw_wr_hdr *wr;
 		unsigned int ndesc = skb->priority;     /* previously saved */
 
-		/*
-		 * Write descriptors and free skbs outside the lock to limit
+		written += ndesc;
+		/* Write descriptors and free skbs outside the lock to limit
 		 * wait times.  q->full is still set so new skbs will be queued.
 		 */
+		wr = (struct fw_wr_hdr *)&q->q.desc[q->q.pidx];
+		txq_advance(&q->q, ndesc);
 		spin_unlock(&q->sendq.lock);
 
-		wr = (struct fw_wr_hdr *)&q->q.desc[q->q.pidx];
 		inline_tx_skb(skb, &q->q, wr);
 		kfree_skb(skb);
 
-		written += ndesc;
-		txq_advance(&q->q, ndesc);
 		if (unlikely(txq_avail(&q->q) < TXQ_STOP_THRES)) {
 			unsigned long old = q->q.stops;
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
index 2b52aae7ec86..44806253c178 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
@@ -37,6 +37,7 @@
 #include "t4_regs.h"
 #include "t4_values.h"
 #include "t4fw_api.h"
+#include "t4fw_version.h"
 
 /**
  *	t4_wait_op_done_val - wait until an operation is completed
@@ -345,6 +346,43 @@ int t4_wr_mbox_meat(struct adapter *adap, int mbox, const void *cmd, int size,
 				       FW_CMD_MAX_TIMEOUT);
 }
 
+static int t4_edc_err_read(struct adapter *adap, int idx)
+{
+	u32 edc_ecc_err_addr_reg;
+	u32 rdata_reg;
+
+	if (is_t4(adap->params.chip)) {
+		CH_WARN(adap, "%s: T4 NOT supported.\n", __func__);
+		return 0;
+	}
+	if (idx != 0 && idx != 1) {
+		CH_WARN(adap, "%s: idx %d NOT supported.\n", __func__, idx);
+		return 0;
+	}
+
+	edc_ecc_err_addr_reg = EDC_T5_REG(EDC_H_ECC_ERR_ADDR_A, idx);
+	rdata_reg = EDC_T5_REG(EDC_H_BIST_STATUS_RDATA_A, idx);
+
+	CH_WARN(adap,
+		"edc%d err addr 0x%x: 0x%x.\n",
+		idx, edc_ecc_err_addr_reg,
+		t4_read_reg(adap, edc_ecc_err_addr_reg));
+	CH_WARN(adap,
+		"bist: 0x%x, status %llx %llx %llx %llx %llx %llx %llx %llx %llx.\n",
+		rdata_reg,
+		(unsigned long long)t4_read_reg64(adap, rdata_reg),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 8),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 16),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 24),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 32),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 40),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 48),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 56),
+		(unsigned long long)t4_read_reg64(adap, rdata_reg + 64));
+
+	return 0;
+}
+
 /**
  *	t4_memory_rw - read/write EDC 0, EDC 1 or MC via PCIE memory window
  *	@adap: the adapter
@@ -1322,9 +1360,10 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 	};
 
 	static const unsigned int t6_reg_ranges[] = {
-		0x1008, 0x114c,
+		0x1008, 0x1124,
+		0x1138, 0x114c,
 		0x1180, 0x11b4,
-		0x11fc, 0x1250,
+		0x11fc, 0x1254,
 		0x1280, 0x133c,
 		0x1800, 0x18fc,
 		0x3000, 0x302c,
@@ -1345,18 +1384,18 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x5a80, 0x5a9c,
 		0x5b94, 0x5bfc,
 		0x5c10, 0x5ec0,
-		0x5ec8, 0x5ec8,
+		0x5ec8, 0x5ecc,
 		0x6000, 0x6040,
-		0x6058, 0x6154,
+		0x6058, 0x619c,
 		0x7700, 0x7798,
 		0x77c0, 0x7880,
 		0x78cc, 0x78fc,
 		0x7b00, 0x7c54,
 		0x7d00, 0x7efc,
-		0x8dc0, 0x8de0,
+		0x8dc0, 0x8de4,
 		0x8df8, 0x8e84,
 		0x8ea0, 0x8f88,
-		0x8fb8, 0x911c,
+		0x8fb8, 0x9124,
 		0x9400, 0x9470,
 		0x9600, 0x971c,
 		0x9800, 0x9808,
@@ -1371,20 +1410,21 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x9f00, 0x9f6c,
 		0x9f80, 0xa020,
 		0xd004, 0xd03c,
+		0xd100, 0xd118,
+		0xd200, 0xd31c,
 		0xdfc0, 0xdfe0,
 		0xe000, 0xf008,
 		0x11000, 0x11014,
-		0x11048, 0x11110,
-		0x11118, 0x1117c,
-		0x11190, 0x11260,
+		0x11048, 0x1117c,
+		0x11190, 0x11270,
 		0x11300, 0x1130c,
-		0x12000, 0x1205c,
+		0x12000, 0x1206c,
 		0x19040, 0x1906c,
 		0x19078, 0x19080,
 		0x1908c, 0x19124,
 		0x19150, 0x191b0,
 		0x191d0, 0x191e8,
-		0x19238, 0x192b8,
+		0x19238, 0x192bc,
 		0x193f8, 0x19474,
 		0x19490, 0x194cc,
 		0x194f0, 0x194f8,
@@ -1461,12 +1501,11 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x1ff00, 0x1ff84,
 		0x1ffc0, 0x1ffc8,
 		0x30000, 0x30070,
-		0x30100, 0x3015c,
-		0x30190, 0x301d0,
-		0x30200, 0x30318,
+		0x30100, 0x301d0,
+		0x30200, 0x30320,
 		0x30400, 0x3052c,
 		0x30540, 0x3061c,
-		0x30800, 0x3088c,
+		0x30800, 0x30890,
 		0x308c0, 0x30908,
 		0x30910, 0x309b8,
 		0x30a00, 0x30a04,
@@ -1539,12 +1578,11 @@ void t4_get_regs(struct adapter *adap, void *buf, size_t buf_size)
 		0x33c24, 0x33c50,
 		0x33cf0, 0x33cfc,
 		0x34000, 0x34070,
-		0x34100, 0x3415c,
-		0x34190, 0x341d0,
-		0x34200, 0x34318,
+		0x34100, 0x341d0,
+		0x34200, 0x34320,
 		0x34400, 0x3452c,
 		0x34540, 0x3461c,
-		0x34800, 0x3488c,
+		0x34800, 0x34890,
 		0x348c0, 0x34908,
 		0x34910, 0x349b8,
 		0x34a00, 0x34a04,
@@ -2129,6 +2167,61 @@ int t4_get_exprom_version(struct adapter *adap, u32 *vers)
 	return 0;
 }
 
+/**
+ *	t4_check_fw_version - check if the FW is supported with this driver
+ *	@adap: the adapter
+ *
+ *	Checks if an adapter's FW is compatible with the driver.  Returns 0
+ *	if there's exact match, a negative error if the version could not be
+ *	read or there's a major version mismatch
+ */
+int t4_check_fw_version(struct adapter *adap)
+{
+	int ret, major, minor, micro;
+	int exp_major, exp_minor, exp_micro;
+	unsigned int chip_version = CHELSIO_CHIP_VERSION(adap->params.chip);
+
+	ret = t4_get_fw_version(adap, &adap->params.fw_vers);
+	if (ret)
+		return ret;
+
+	major = FW_HDR_FW_VER_MAJOR_G(adap->params.fw_vers);
+	minor = FW_HDR_FW_VER_MINOR_G(adap->params.fw_vers);
+	micro = FW_HDR_FW_VER_MICRO_G(adap->params.fw_vers);
+
+	switch (chip_version) {
+	case CHELSIO_T4:
+		exp_major = T4FW_MIN_VERSION_MAJOR;
+		exp_minor = T4FW_MIN_VERSION_MINOR;
+		exp_micro = T4FW_MIN_VERSION_MICRO;
+		break;
+	case CHELSIO_T5:
+		exp_major = T5FW_MIN_VERSION_MAJOR;
+		exp_minor = T5FW_MIN_VERSION_MINOR;
+		exp_micro = T5FW_MIN_VERSION_MICRO;
+		break;
+	case CHELSIO_T6:
+		exp_major = T6FW_MIN_VERSION_MAJOR;
+		exp_minor = T6FW_MIN_VERSION_MINOR;
+		exp_micro = T6FW_MIN_VERSION_MICRO;
+		break;
+	default:
+		dev_err(adap->pdev_dev, "Unsupported chip type, %x\n",
+			adap->chip);
+		return -EINVAL;
+	}
+
+	if (major < exp_major || (major == exp_major && minor < exp_minor) ||
+	    (major == exp_major && minor == exp_minor && micro < exp_micro)) {
+		dev_err(adap->pdev_dev,
+			"Card has firmware version %u.%u.%u, minimum "
+			"supported firmware is %u.%u.%u.\n", major, minor,
+			micro, exp_major, exp_minor, exp_micro);
+		return -EFAULT;
+	}
+	return 0;
+}
+
 /* Is the given firmware API compatible with the one the driver was compiled
  * with?
  */
@@ -3281,6 +3374,8 @@ static void mem_intr_handler(struct adapter *adapter, int idx)
 	if (v & ECC_CE_INT_CAUSE_F) {
 		u32 cnt = ECC_CECNT_G(t4_read_reg(adapter, cnt_addr));
 
+		t4_edc_err_read(adapter, idx);
+
 		t4_write_reg(adapter, cnt_addr, ECC_CECNT_V(ECC_CECNT_M));
 		if (printk_ratelimit())
 			dev_warn(adapter->pdev_dev,
@@ -3488,7 +3583,9 @@ int t4_slow_intr_handler(struct adapter *adapter)
 void t4_intr_enable(struct adapter *adapter)
 {
 	u32 val = 0;
-	u32 pf = SOURCEPF_G(t4_read_reg(adapter, PL_WHOAMI_A));
+	u32 whoami = t4_read_reg(adapter, PL_WHOAMI_A);
+	u32 pf = CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5 ?
+			SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami);
 
 	if (CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5)
 		val = ERR_DROPPED_DB_F | ERR_EGR_CTXT_PRIO_F | DBFIFO_HP_INT_F;
@@ -3513,7 +3610,9 @@ void t4_intr_enable(struct adapter *adapter)
  */
 void t4_intr_disable(struct adapter *adapter)
 {
-	u32 pf = SOURCEPF_G(t4_read_reg(adapter, PL_WHOAMI_A));
+	u32 whoami = t4_read_reg(adapter, PL_WHOAMI_A);
+	u32 pf = CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5 ?
+			SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami);
 
 	t4_write_reg(adapter, MYPF_REG(PL_PF_INT_ENABLE_A), 0);
 	t4_set_reg_field(adapter, PL_INT_MAP0_A, 1 << pf, 0);
@@ -3687,6 +3786,11 @@ int t4_read_rss(struct adapter *adapter, u16 *map)
 	return 0;
 }
 
+static unsigned int t4_use_ldst(struct adapter *adap)
+{
+	return (adap->flags & FW_OK) || !adap->use_bd;
+}
+
 /**
  *	t4_fw_tp_pio_rw - Access TP PIO through LDST
  *	@adap: the adapter
@@ -3730,7 +3834,7 @@ static void t4_fw_tp_pio_rw(struct adapter *adap, u32 *vals, unsigned int nregs,
  */
 void t4_read_rss_key(struct adapter *adap, u32 *key)
 {
-	if (adap->flags & FW_OK)
+	if (t4_use_ldst(adap))
 		t4_fw_tp_pio_rw(adap, key, 10, TP_RSS_SECRET_KEY0_A, 1);
 	else
 		t4_read_indirect(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A, key, 10,
@@ -3760,7 +3864,7 @@ void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx)
 	    (vrt & KEYEXTEND_F) && (KEYMODE_G(vrt) == 3))
 		rss_key_addr_cnt = 32;
 
-	if (adap->flags & FW_OK)
+	if (t4_use_ldst(adap))
 		t4_fw_tp_pio_rw(adap, (void *)key, 10, TP_RSS_SECRET_KEY0_A, 0);
 	else
 		t4_write_indirect(adap, TP_PIO_ADDR_A, TP_PIO_DATA_A, key, 10,
@@ -3789,7 +3893,7 @@ void t4_write_rss_key(struct adapter *adap, const u32 *key, int idx)
 void t4_read_rss_pf_config(struct adapter *adapter, unsigned int index,
 			   u32 *valp)
 {
-	if (adapter->flags & FW_OK)
+	if (t4_use_ldst(adapter))
 		t4_fw_tp_pio_rw(adapter, valp, 1,
 				TP_RSS_PF0_CONFIG_A + index, 1);
 	else
@@ -3829,7 +3933,7 @@ void t4_read_rss_vf_config(struct adapter *adapter, unsigned int index,
 
 	/* Grab the VFL/VFH values ...
 	 */
-	if (adapter->flags & FW_OK) {
+	if (t4_use_ldst(adapter)) {
 		t4_fw_tp_pio_rw(adapter, vfl, 1, TP_RSS_VFL_CONFIG_A, 1);
 		t4_fw_tp_pio_rw(adapter, vfh, 1, TP_RSS_VFH_CONFIG_A, 1);
 	} else {
@@ -3850,7 +3954,7 @@ u32 t4_read_rss_pf_map(struct adapter *adapter)
 {
 	u32 pfmap;
 
-	if (adapter->flags & FW_OK)
+	if (t4_use_ldst(adapter))
 		t4_fw_tp_pio_rw(adapter, &pfmap, 1, TP_RSS_PF_MAP_A, 1);
 	else
 		t4_read_indirect(adapter, TP_PIO_ADDR_A, TP_PIO_DATA_A,
@@ -3868,7 +3972,7 @@ u32 t4_read_rss_pf_mask(struct adapter *adapter)
 {
 	u32 pfmask;
 
-	if (adapter->flags & FW_OK)
+	if (t4_use_ldst(adapter))
 		t4_fw_tp_pio_rw(adapter, &pfmask, 1, TP_RSS_PF_MSK_A, 1);
 	else
 		t4_read_indirect(adapter, TP_PIO_ADDR_A, TP_PIO_DATA_A,
@@ -3924,43 +4028,25 @@ void t4_tp_get_tcp_stats(struct adapter *adap, struct tp_tcp_stats *v4,
  */
 void t4_tp_get_err_stats(struct adapter *adap, struct tp_err_stats *st)
 {
-	/* T6 and later has 2 channels */
-	if (adap->params.arch.nchan == NCHAN) {
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->mac_in_errs, 12, TP_MIB_MAC_IN_ERR_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->tnl_cong_drops, 8,
-				 TP_MIB_TNL_CNG_DROP_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->tnl_tx_drops, 4,
-				 TP_MIB_TNL_DROP_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->ofld_vlan_drops, 4,
-				 TP_MIB_OFD_VLN_DROP_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->tcp6_in_errs, 4,
-				 TP_MIB_TCP_V6IN_ERR_0_A);
-	} else {
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->mac_in_errs, 2, TP_MIB_MAC_IN_ERR_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->hdr_in_errs, 2, TP_MIB_HDR_IN_ERR_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->tcp_in_errs, 2, TP_MIB_TCP_IN_ERR_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->tnl_cong_drops, 2,
-				 TP_MIB_TNL_CNG_DROP_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->ofld_chan_drops, 2,
-				 TP_MIB_OFD_CHN_DROP_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->tnl_tx_drops, 2, TP_MIB_TNL_DROP_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->ofld_vlan_drops, 2,
-				 TP_MIB_OFD_VLN_DROP_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
-				 st->tcp6_in_errs, 2, TP_MIB_TCP_V6IN_ERR_0_A);
-	}
+	int nchan = adap->params.arch.nchan;
+
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->mac_in_errs, nchan, TP_MIB_MAC_IN_ERR_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->hdr_in_errs, nchan, TP_MIB_HDR_IN_ERR_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->tcp_in_errs, nchan, TP_MIB_TCP_IN_ERR_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->tnl_cong_drops, nchan, TP_MIB_TNL_CNG_DROP_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->ofld_chan_drops, nchan, TP_MIB_OFD_CHN_DROP_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->tnl_tx_drops, nchan, TP_MIB_TNL_DROP_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->ofld_vlan_drops, nchan, TP_MIB_OFD_VLN_DROP_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
+			 st->tcp6_in_errs, nchan, TP_MIB_TCP_V6IN_ERR_0_A);
+
 	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A,
 			 &st->ofld_no_neigh, 2, TP_MIB_OFD_ARP_DROP_A);
 }
@@ -3974,16 +4060,13 @@ void t4_tp_get_err_stats(struct adapter *adap, struct tp_err_stats *st)
  */
 void t4_tp_get_cpl_stats(struct adapter *adap, struct tp_cpl_stats *st)
 {
-	/* T6 and later has 2 channels */
-	if (adap->params.arch.nchan == NCHAN) {
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, st->req,
-				 8, TP_MIB_CPL_IN_REQ_0_A);
-	} else {
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, st->req,
-				 2, TP_MIB_CPL_IN_REQ_0_A);
-		t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, st->rsp,
-				 2, TP_MIB_CPL_OUT_RSP_0_A);
-	}
+	int nchan = adap->params.arch.nchan;
+
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, st->req,
+			 nchan, TP_MIB_CPL_IN_REQ_0_A);
+	t4_read_indirect(adap, TP_MIB_INDEX_A, TP_MIB_DATA_A, st->rsp,
+			 nchan, TP_MIB_CPL_OUT_RSP_0_A);
+
 }
 
 /**
@@ -4238,6 +4321,119 @@ void t4_get_chan_txrate(struct adapter *adap, u64 *nic_rate, u64 *ofld_rate)
 }
 
 /**
+ *	t4_set_trace_filter - configure one of the tracing filters
+ *	@adap: the adapter
+ *	@tp: the desired trace filter parameters
+ *	@idx: which filter to configure
+ *	@enable: whether to enable or disable the filter
+ *
+ *	Configures one of the tracing filters available in HW.  If @enable is
+ *	%0 @tp is not examined and may be %NULL. The user is responsible to
+ *	set the single/multiple trace mode by writing to MPS_TRC_CFG_A register
+ */
+int t4_set_trace_filter(struct adapter *adap, const struct trace_params *tp,
+			int idx, int enable)
+{
+	int i, ofst = idx * 4;
+	u32 data_reg, mask_reg, cfg;
+	u32 multitrc = TRCMULTIFILTER_F;
+
+	if (!enable) {
+		t4_write_reg(adap, MPS_TRC_FILTER_MATCH_CTL_A_A + ofst, 0);
+		return 0;
+	}
+
+	cfg = t4_read_reg(adap, MPS_TRC_CFG_A);
+	if (cfg & TRCMULTIFILTER_F) {
+		/* If multiple tracers are enabled, then maximum
+		 * capture size is 2.5KB (FIFO size of a single channel)
+		 * minus 2 flits for CPL_TRACE_PKT header.
+		 */
+		if (tp->snap_len > ((10 * 1024 / 4) - (2 * 8)))
+			return -EINVAL;
+	} else {
+		/* If multiple tracers are disabled, to avoid deadlocks
+		 * maximum packet capture size of 9600 bytes is recommended.
+		 * Also in this mode, only trace0 can be enabled and running.
+		 */
+		multitrc = 0;
+		if (tp->snap_len > 9600 || idx)
+			return -EINVAL;
+	}
+
+	if (tp->port > (is_t4(adap->params.chip) ? 11 : 19) || tp->invert > 1 ||
+	    tp->skip_len > TFLENGTH_M || tp->skip_ofst > TFOFFSET_M ||
+	    tp->min_len > TFMINPKTSIZE_M)
+		return -EINVAL;
+
+	/* stop the tracer we'll be changing */
+	t4_write_reg(adap, MPS_TRC_FILTER_MATCH_CTL_A_A + ofst, 0);
+
+	idx *= (MPS_TRC_FILTER1_MATCH_A - MPS_TRC_FILTER0_MATCH_A);
+	data_reg = MPS_TRC_FILTER0_MATCH_A + idx;
+	mask_reg = MPS_TRC_FILTER0_DONT_CARE_A + idx;
+
+	for (i = 0; i < TRACE_LEN / 4; i++, data_reg += 4, mask_reg += 4) {
+		t4_write_reg(adap, data_reg, tp->data[i]);
+		t4_write_reg(adap, mask_reg, ~tp->mask[i]);
+	}
+	t4_write_reg(adap, MPS_TRC_FILTER_MATCH_CTL_B_A + ofst,
+		     TFCAPTUREMAX_V(tp->snap_len) |
+		     TFMINPKTSIZE_V(tp->min_len));
+	t4_write_reg(adap, MPS_TRC_FILTER_MATCH_CTL_A_A + ofst,
+		     TFOFFSET_V(tp->skip_ofst) | TFLENGTH_V(tp->skip_len) |
+		     (is_t4(adap->params.chip) ?
+		     TFPORT_V(tp->port) | TFEN_F | TFINVERTMATCH_V(tp->invert) :
+		     T5_TFPORT_V(tp->port) | T5_TFEN_F |
+		     T5_TFINVERTMATCH_V(tp->invert)));
+
+	return 0;
+}
+
+/**
+ *	t4_get_trace_filter - query one of the tracing filters
+ *	@adap: the adapter
+ *	@tp: the current trace filter parameters
+ *	@idx: which trace filter to query
+ *	@enabled: non-zero if the filter is enabled
+ *
+ *	Returns the current settings of one of the HW tracing filters.
+ */
+void t4_get_trace_filter(struct adapter *adap, struct trace_params *tp, int idx,
+			 int *enabled)
+{
+	u32 ctla, ctlb;
+	int i, ofst = idx * 4;
+	u32 data_reg, mask_reg;
+
+	ctla = t4_read_reg(adap, MPS_TRC_FILTER_MATCH_CTL_A_A + ofst);
+	ctlb = t4_read_reg(adap, MPS_TRC_FILTER_MATCH_CTL_B_A + ofst);
+
+	if (is_t4(adap->params.chip)) {
+		*enabled = !!(ctla & TFEN_F);
+		tp->port =  TFPORT_G(ctla);
+		tp->invert = !!(ctla & TFINVERTMATCH_F);
+	} else {
+		*enabled = !!(ctla & T5_TFEN_F);
+		tp->port = T5_TFPORT_G(ctla);
+		tp->invert = !!(ctla & T5_TFINVERTMATCH_F);
+	}
+	tp->snap_len = TFCAPTUREMAX_G(ctlb);
+	tp->min_len = TFMINPKTSIZE_G(ctlb);
+	tp->skip_ofst = TFOFFSET_G(ctla);
+	tp->skip_len = TFLENGTH_G(ctla);
+
+	ofst = (MPS_TRC_FILTER1_MATCH_A - MPS_TRC_FILTER0_MATCH_A) * idx;
+	data_reg = MPS_TRC_FILTER0_MATCH_A + ofst;
+	mask_reg = MPS_TRC_FILTER0_DONT_CARE_A + ofst;
+
+	for (i = 0; i < TRACE_LEN / 4; i++, data_reg += 4, mask_reg += 4) {
+		tp->mask[i] = ~t4_read_reg(adap, mask_reg);
+		tp->data[i] = t4_read_reg(adap, data_reg) & tp->mask[i];
+	}
+}
+
+/**
  *	t4_pmtx_get_stats - returns the HW stats from PMTX
  *	@adap: the adapter
  *	@cnt: where to store the count statistics
@@ -6294,7 +6490,7 @@ int t4_init_tp_params(struct adapter *adap)
 	/* Cache the adapter's Compressed Filter Mode and global Incress
 	 * Configuration.
 	 */
-	if (adap->flags & FW_OK) {
+	if (t4_use_ldst(adap)) {
 		t4_fw_tp_pio_rw(adap, &adap->params.tp.vlan_pri_map, 1,
 				TP_VLAN_PRI_MAP_A, 1);
 		t4_fw_tp_pio_rw(adap, &adap->params.tp.ingress_config, 1,
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
index c8488f430d19..640369df8b3a 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.h
@@ -47,7 +47,6 @@ enum {
 	TCB_SIZE       = 128,   /* TCB size */
 	NMTUS          = 16,    /* size of MTU table */
 	NCCTRL_WIN     = 32,    /* # of congestion control windows */
-	L2T_SIZE       = 4096,  /* # of L2T entries */
 	PM_NSTATS      = 5,     /* # of PM stats */
 	MBOX_LEN       = 64,    /* mailbox size in bytes */
 	TRACE_LEN      = 112,   /* length of trace data and mask */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
index 132cb8fc0bf7..b99144afd4ec 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h
@@ -660,6 +660,9 @@ struct cpl_tx_pkt {
 #define TXPKT_OVLAN_IDX_S    12
 #define TXPKT_OVLAN_IDX_V(x) ((x) << TXPKT_OVLAN_IDX_S)
 
+#define TXPKT_T5_OVLAN_IDX_S	12
+#define TXPKT_T5_OVLAN_IDX_V(x)	((x) << TXPKT_T5_OVLAN_IDX_S)
+
 #define TXPKT_INTF_S    16
 #define TXPKT_INTF_V(x) ((x) << TXPKT_INTF_S)
 
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
index d7ca106927b0..8353a6cbfcc2 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
@@ -142,6 +142,8 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN
 	CH_PCI_ID_TABLE_FENTRY(0x5013),	/* T580-chr */
 	CH_PCI_ID_TABLE_FENTRY(0x5014),	/* T580-so */
 	CH_PCI_ID_TABLE_FENTRY(0x5015),	/* T502-bt */
+	CH_PCI_ID_TABLE_FENTRY(0x5016),	/* T580-OCP-SO */
+	CH_PCI_ID_TABLE_FENTRY(0x5017),	/* T520-OCP-SO */
 	CH_PCI_ID_TABLE_FENTRY(0x5080),	/* Custom T540-cr */
 	CH_PCI_ID_TABLE_FENTRY(0x5081),	/* Custom T540-LL-cr */
 	CH_PCI_ID_TABLE_FENTRY(0x5082),	/* Custom T504-cr */
@@ -155,6 +157,22 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN
 	CH_PCI_ID_TABLE_FENTRY(0x5090),	/* Custom T540-CR */
 	CH_PCI_ID_TABLE_FENTRY(0x5091),	/* Custom T522-CR */
 	CH_PCI_ID_TABLE_FENTRY(0x5092),	/* Custom T520-CR */
+
+	/* T6 adapters:
+	 */
+	CH_PCI_ID_TABLE_FENTRY(0x6001),
+	CH_PCI_ID_TABLE_FENTRY(0x6002),
+	CH_PCI_ID_TABLE_FENTRY(0x6003),
+	CH_PCI_ID_TABLE_FENTRY(0x6004),
+	CH_PCI_ID_TABLE_FENTRY(0x6005),
+	CH_PCI_ID_TABLE_FENTRY(0x6006),
+	CH_PCI_ID_TABLE_FENTRY(0x6007),
+	CH_PCI_ID_TABLE_FENTRY(0x6009),
+	CH_PCI_ID_TABLE_FENTRY(0x600d),
+	CH_PCI_ID_TABLE_FENTRY(0x6010),
+	CH_PCI_ID_TABLE_FENTRY(0x6011),
+	CH_PCI_ID_TABLE_FENTRY(0x6014),
+	CH_PCI_ID_TABLE_FENTRY(0x6015),
 CH_PCI_DEVICE_ID_TABLE_DEFINE_END;
 
 #endif /* __T4_PCI_ID_TBL_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
index 375a825573b0..fc3044c8ac1c 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
@@ -136,6 +136,20 @@
 #define  INGPACKBOUNDARY_G(x)	(((x) >> INGPACKBOUNDARY_S) \
 				 & INGPACKBOUNDARY_M)
 
+#define VFIFO_ENABLE_S    10
+#define VFIFO_ENABLE_V(x) ((x) << VFIFO_ENABLE_S)
+#define VFIFO_ENABLE_F    VFIFO_ENABLE_V(1U)
+
+#define SGE_DBVFIFO_BADDR_A 0x1138
+
+#define DBVFIFO_SIZE_S    6
+#define DBVFIFO_SIZE_M    0xfffU
+#define DBVFIFO_SIZE_G(x) (((x) >> DBVFIFO_SIZE_S) & DBVFIFO_SIZE_M)
+
+#define T6_DBVFIFO_SIZE_S    0
+#define T6_DBVFIFO_SIZE_M    0x1fffU
+#define T6_DBVFIFO_SIZE_G(x) (((x) >> T6_DBVFIFO_SIZE_S) & T6_DBVFIFO_SIZE_M)
+
 #define GLOBALENABLE_S    0
 #define GLOBALENABLE_V(x) ((x) << GLOBALENABLE_S)
 #define GLOBALENABLE_F    GLOBALENABLE_V(1U)
@@ -303,6 +317,8 @@
 #define SGE_FL_BUFFER_SIZE7_A 0x1060
 #define SGE_FL_BUFFER_SIZE8_A 0x1064
 
+#define SGE_IMSG_CTXT_BADDR_A 0x1088
+#define SGE_FLM_CACHE_BADDR_A 0x108c
 #define SGE_INGRESS_RX_THRESHOLD_A 0x10a0
 
 #define THRESHOLD_0_S    24
@@ -338,6 +354,11 @@
 #define EGRTHRESHOLDPACKING_G(x) \
 	(((x) >> EGRTHRESHOLDPACKING_S) & EGRTHRESHOLDPACKING_M)
 
+#define T6_EGRTHRESHOLDPACKING_S    16
+#define T6_EGRTHRESHOLDPACKING_M    0xffU
+#define T6_EGRTHRESHOLDPACKING_G(x) \
+	(((x) >> T6_EGRTHRESHOLDPACKING_S) & T6_EGRTHRESHOLDPACKING_M)
+
 #define SGE_TIMESTAMP_LO_A 0x1098
 #define SGE_TIMESTAMP_HI_A 0x109c
 
@@ -352,6 +373,7 @@
 #define TSVAL_G(x) (((x) >> TSVAL_S) & TSVAL_M)
 
 #define SGE_DBFIFO_STATUS_A 0x10a4
+#define SGE_DBVFIFO_SIZE_A 0x113c
 
 #define HP_INT_THRESH_S    28
 #define HP_INT_THRESH_M    0xfU
@@ -864,6 +886,10 @@
 /* registers for module MA */
 #define MA_EDRAM0_BAR_A 0x77c0
 
+#define EDRAM0_BASE_S    16
+#define EDRAM0_BASE_M    0xfffU
+#define EDRAM0_BASE_G(x) (((x) >> EDRAM0_BASE_S) & EDRAM0_BASE_M)
+
 #define EDRAM0_SIZE_S    0
 #define EDRAM0_SIZE_M    0xfffU
 #define EDRAM0_SIZE_V(x) ((x) << EDRAM0_SIZE_S)
@@ -871,6 +897,10 @@
 
 #define MA_EDRAM1_BAR_A 0x77c4
 
+#define EDRAM1_BASE_S    16
+#define EDRAM1_BASE_M    0xfffU
+#define EDRAM1_BASE_G(x) (((x) >> EDRAM1_BASE_S) & EDRAM1_BASE_M)
+
 #define EDRAM1_SIZE_S    0
 #define EDRAM1_SIZE_M    0xfffU
 #define EDRAM1_SIZE_V(x) ((x) << EDRAM1_SIZE_S)
@@ -878,6 +908,11 @@
 
 #define MA_EXT_MEMORY_BAR_A 0x77c8
 
+#define EXT_MEM_BASE_S    16
+#define EXT_MEM_BASE_M    0xfffU
+#define EXT_MEM_BASE_V(x) ((x) << EXT_MEM_BASE_S)
+#define EXT_MEM_BASE_G(x) (((x) >> EXT_MEM_BASE_S) & EXT_MEM_BASE_M)
+
 #define EXT_MEM_SIZE_S    0
 #define EXT_MEM_SIZE_M    0xfffU
 #define EXT_MEM_SIZE_V(x) ((x) << EXT_MEM_SIZE_S)
@@ -885,6 +920,10 @@
 
 #define MA_EXT_MEMORY1_BAR_A 0x7808
 
+#define EXT_MEM1_BASE_S    16
+#define EXT_MEM1_BASE_M    0xfffU
+#define EXT_MEM1_BASE_G(x) (((x) >> EXT_MEM1_BASE_S) & EXT_MEM1_BASE_M)
+
 #define EXT_MEM1_SIZE_S    0
 #define EXT_MEM1_SIZE_M    0xfffU
 #define EXT_MEM1_SIZE_V(x) ((x) << EXT_MEM1_SIZE_S)
@@ -892,6 +931,10 @@
 
 #define MA_EXT_MEMORY0_BAR_A 0x77c8
 
+#define EXT_MEM0_BASE_S    16
+#define EXT_MEM0_BASE_M    0xfffU
+#define EXT_MEM0_BASE_G(x) (((x) >> EXT_MEM0_BASE_S) & EXT_MEM0_BASE_M)
+
 #define EXT_MEM0_SIZE_S    0
 #define EXT_MEM0_SIZE_M    0xfffU
 #define EXT_MEM0_SIZE_V(x) ((x) << EXT_MEM0_SIZE_S)
@@ -973,6 +1016,10 @@
 
 /* registers for module CIM */
 #define CIM_BOOT_CFG_A 0x7b00
+#define CIM_SDRAM_BASE_ADDR_A 0x7b14
+#define CIM_SDRAM_ADDR_SIZE_A 0x7b18
+#define CIM_EXTMEM2_BASE_ADDR_A 0x7b1c
+#define CIM_EXTMEM2_ADDR_SIZE_A 0x7b20
 #define CIM_PF_MAILBOX_CTRL_SHADOW_COPY_A 0x290
 
 #define  BOOTADDR_M	0xffffff00U
@@ -1231,6 +1278,33 @@
 #define TP_OUT_CONFIG_A		0x7d04
 #define TP_GLOBAL_CONFIG_A	0x7d08
 
+#define TP_CMM_TCB_BASE_A 0x7d10
+#define TP_CMM_MM_BASE_A 0x7d14
+#define TP_CMM_TIMER_BASE_A 0x7d18
+#define TP_PMM_TX_BASE_A 0x7d20
+#define TP_PMM_RX_BASE_A 0x7d28
+#define TP_PMM_RX_PAGE_SIZE_A 0x7d2c
+#define TP_PMM_RX_MAX_PAGE_A 0x7d30
+#define TP_PMM_TX_PAGE_SIZE_A 0x7d34
+#define TP_PMM_TX_MAX_PAGE_A 0x7d38
+#define TP_CMM_MM_MAX_PSTRUCT_A 0x7e6c
+
+#define PMRXNUMCHN_S    31
+#define PMRXNUMCHN_V(x) ((x) << PMRXNUMCHN_S)
+#define PMRXNUMCHN_F    PMRXNUMCHN_V(1U)
+
+#define PMTXNUMCHN_S    30
+#define PMTXNUMCHN_M    0x3U
+#define PMTXNUMCHN_G(x) (((x) >> PMTXNUMCHN_S) & PMTXNUMCHN_M)
+
+#define PMTXMAXPAGE_S    0
+#define PMTXMAXPAGE_M    0x1fffffU
+#define PMTXMAXPAGE_G(x) (((x) >> PMTXMAXPAGE_S) & PMTXMAXPAGE_M)
+
+#define PMRXMAXPAGE_S    0
+#define PMRXMAXPAGE_M    0x1fffffU
+#define PMRXMAXPAGE_G(x) (((x) >> PMRXMAXPAGE_S) & PMRXMAXPAGE_M)
+
 #define DBGLAMODE_S	14
 #define DBGLAMODE_M	0x3U
 #define DBGLAMODE_G(x)	(((x) >> DBGLAMODE_S) & DBGLAMODE_M)
@@ -1338,6 +1412,9 @@
 #define MTUVALUE_G(x) (((x) >> MTUVALUE_S) & MTUVALUE_M)
 
 #define TP_RSS_LKP_TABLE_A	0x7dec
+#define TP_CMM_MM_RX_FLST_BASE_A 0x7e60
+#define TP_CMM_MM_TX_FLST_BASE_A 0x7e64
+#define TP_CMM_MM_PS_FLST_BASE_A 0x7e68
 
 #define LKPTBLROWVLD_S    31
 #define LKPTBLROWVLD_V(x) ((x) << LKPTBLROWVLD_S)
@@ -1483,6 +1560,11 @@
 #define TP_MIB_RQE_DFR_PKT_A	0x64
 
 #define ULP_TX_INT_CAUSE_A	0x8dcc
+#define ULP_TX_TPT_LLIMIT_A	0x8dd4
+#define ULP_TX_TPT_ULIMIT_A	0x8dd8
+#define ULP_TX_PBL_LLIMIT_A	0x8ddc
+#define ULP_TX_PBL_ULIMIT_A	0x8de0
+#define ULP_TX_ERR_TABLE_BASE_A 0x8e04
 
 #define PBL_BOUND_ERR_CH3_S    31
 #define PBL_BOUND_ERR_CH3_V(x) ((x) << PBL_BOUND_ERR_CH3_S)
@@ -1804,6 +1886,9 @@
 #define TRCMULTIFILTER_F    TRCMULTIFILTER_V(1U)
 
 #define MPS_TRC_RSS_CONTROL_A		0x9808
+#define MPS_TRC_FILTER1_RSS_CONTROL_A	0x9ff4
+#define MPS_TRC_FILTER2_RSS_CONTROL_A	0x9ffc
+#define MPS_TRC_FILTER3_RSS_CONTROL_A	0xa004
 #define MPS_T5_TRC_RSS_CONTROL_A	0xa00c
 
 #define RSSCONTROL_S    16
@@ -1812,6 +1897,59 @@
 #define QUEUENUMBER_S    0
 #define QUEUENUMBER_V(x) ((x) << QUEUENUMBER_S)
 
+#define TFINVERTMATCH_S    24
+#define TFINVERTMATCH_V(x) ((x) << TFINVERTMATCH_S)
+#define TFINVERTMATCH_F    TFINVERTMATCH_V(1U)
+
+#define TFEN_S    22
+#define TFEN_V(x) ((x) << TFEN_S)
+#define TFEN_F    TFEN_V(1U)
+
+#define TFPORT_S    18
+#define TFPORT_M    0xfU
+#define TFPORT_V(x) ((x) << TFPORT_S)
+#define TFPORT_G(x) (((x) >> TFPORT_S) & TFPORT_M)
+
+#define TFLENGTH_S    8
+#define TFLENGTH_M    0x1fU
+#define TFLENGTH_V(x) ((x) << TFLENGTH_S)
+#define TFLENGTH_G(x) (((x) >> TFLENGTH_S) & TFLENGTH_M)
+
+#define TFOFFSET_S    0
+#define TFOFFSET_M    0x1fU
+#define TFOFFSET_V(x) ((x) << TFOFFSET_S)
+#define TFOFFSET_G(x) (((x) >> TFOFFSET_S) & TFOFFSET_M)
+
+#define T5_TFINVERTMATCH_S    25
+#define T5_TFINVERTMATCH_V(x) ((x) << T5_TFINVERTMATCH_S)
+#define T5_TFINVERTMATCH_F    T5_TFINVERTMATCH_V(1U)
+
+#define T5_TFEN_S    23
+#define T5_TFEN_V(x) ((x) << T5_TFEN_S)
+#define T5_TFEN_F    T5_TFEN_V(1U)
+
+#define T5_TFPORT_S    18
+#define T5_TFPORT_M    0x1fU
+#define T5_TFPORT_V(x) ((x) << T5_TFPORT_S)
+#define T5_TFPORT_G(x) (((x) >> T5_TFPORT_S) & T5_TFPORT_M)
+
+#define MPS_TRC_FILTER_MATCH_CTL_A_A 0x9810
+#define MPS_TRC_FILTER_MATCH_CTL_B_A 0x9820
+
+#define TFMINPKTSIZE_S    16
+#define TFMINPKTSIZE_M    0x1ffU
+#define TFMINPKTSIZE_V(x) ((x) << TFMINPKTSIZE_S)
+#define TFMINPKTSIZE_G(x) (((x) >> TFMINPKTSIZE_S) & TFMINPKTSIZE_M)
+
+#define TFCAPTUREMAX_S    0
+#define TFCAPTUREMAX_M    0x3fffU
+#define TFCAPTUREMAX_V(x) ((x) << TFCAPTUREMAX_S)
+#define TFCAPTUREMAX_G(x) (((x) >> TFCAPTUREMAX_S) & TFCAPTUREMAX_M)
+
+#define MPS_TRC_FILTER0_MATCH_A 0x9c00
+#define MPS_TRC_FILTER0_DONT_CARE_A 0x9c80
+#define MPS_TRC_FILTER1_MATCH_A 0x9d00
+
 #define TP_RSS_CONFIG_A 0x7df0
 
 #define TNL4TUPENIPV6_S    31
@@ -2247,12 +2385,32 @@
 #define MATCHSRAM_V(x) ((x) << MATCHSRAM_S)
 #define MATCHSRAM_F    MATCHSRAM_V(1U)
 
+#define MPS_RX_PG_RSV0_A 0x11010
+#define MPS_RX_PG_RSV4_A 0x11020
 #define MPS_RX_PERR_INT_CAUSE_A 0x11074
+#define MPS_RX_MAC_BG_PG_CNT0_A 0x11208
+#define MPS_RX_LPBK_BG_PG_CNT0_A 0x11218
 
 #define MPS_CLS_TCAM_Y_L_A 0xf000
 #define MPS_CLS_TCAM_DATA0_A 0xf000
 #define MPS_CLS_TCAM_DATA1_A 0xf004
 
+#define USED_S    16
+#define USED_M    0x7ffU
+#define USED_G(x) (((x) >> USED_S) & USED_M)
+
+#define ALLOC_S    0
+#define ALLOC_M    0x7ffU
+#define ALLOC_G(x) (((x) >> ALLOC_S) & ALLOC_M)
+
+#define T5_USED_S    16
+#define T5_USED_M    0xfffU
+#define T5_USED_G(x) (((x) >> T5_USED_S) & T5_USED_M)
+
+#define T5_ALLOC_S    0
+#define T5_ALLOC_M    0xfffU
+#define T5_ALLOC_G(x) (((x) >> T5_ALLOC_S) & T5_ALLOC_M)
+
 #define DMACH_S    0
 #define DMACH_M    0xffffU
 #define DMACH_G(x) (((x) >> DMACH_S) & DMACH_M)
@@ -2410,8 +2568,21 @@
 #define SLVFIFOPARINT_F    SLVFIFOPARINT_V(1U)
 
 #define ULP_RX_INT_CAUSE_A 0x19158
+#define ULP_RX_ISCSI_LLIMIT_A 0x1915c
+#define ULP_RX_ISCSI_ULIMIT_A 0x19160
 #define ULP_RX_ISCSI_TAGMASK_A 0x19164
 #define ULP_RX_ISCSI_PSZ_A 0x19168
+#define ULP_RX_TDDP_LLIMIT_A 0x1916c
+#define ULP_RX_TDDP_ULIMIT_A 0x19170
+#define ULP_RX_STAG_LLIMIT_A 0x1917c
+#define ULP_RX_STAG_ULIMIT_A 0x19180
+#define ULP_RX_RQ_LLIMIT_A 0x19184
+#define ULP_RX_RQ_ULIMIT_A 0x19188
+#define ULP_RX_PBL_LLIMIT_A 0x1918c
+#define ULP_RX_PBL_ULIMIT_A 0x19190
+#define ULP_RX_CTX_BASE_A 0x19194
+#define ULP_RX_RQUDP_LLIMIT_A 0x191a4
+#define ULP_RX_RQUDP_ULIMIT_A 0x191a8
 #define ULP_RX_LA_CTL_A 0x1923c
 #define ULP_RX_LA_RDPTR_A 0x19240
 #define ULP_RX_LA_RDDATA_A 0x19244
@@ -2473,6 +2644,10 @@
 #define SOURCEPF_M    0x7U
 #define SOURCEPF_G(x) (((x) >> SOURCEPF_S) & SOURCEPF_M)
 
+#define T6_SOURCEPF_S    9
+#define T6_SOURCEPF_M    0x7U
+#define T6_SOURCEPF_G(x) (((x) >> T6_SOURCEPF_S) & T6_SOURCEPF_M)
+
 #define PL_INT_CAUSE_A 0x1940c
 
 #define ULP_TX_S    27
@@ -2612,7 +2787,20 @@
 #define T6_LIPMISS_V(x) ((x) << T6_LIPMISS_S)
 #define T6_LIPMISS_F    T6_LIPMISS_V(1U)
 
+#define LE_DB_CONFIG_A 0x19c04
+#define LE_DB_SERVER_INDEX_A 0x19c18
+#define LE_DB_SRVR_START_INDEX_A 0x19c18
+#define LE_DB_ACT_CNT_IPV4_A 0x19c20
+#define LE_DB_ACT_CNT_IPV6_A 0x19c24
+#define LE_DB_HASH_TID_BASE_A 0x19c30
+#define LE_DB_HASH_TBL_BASE_ADDR_A 0x19c30
 #define LE_DB_INT_CAUSE_A 0x19c3c
+#define LE_DB_TID_HASHBASE_A 0x19df8
+#define T6_LE_DB_HASH_TID_BASE_A 0x19df8
+
+#define HASHEN_S    20
+#define HASHEN_V(x) ((x) << HASHEN_S)
+#define HASHEN_F    HASHEN_V(1U)
 
 #define REQQPARERR_S    16
 #define REQQPARERR_V(x) ((x) << REQQPARERR_S)
@@ -2634,6 +2822,10 @@
 #define LIP0_V(x) ((x) << LIP0_S)
 #define LIP0_F    LIP0_V(1U)
 
+#define BASEADDR_S    3
+#define BASEADDR_M    0x1fffffffU
+#define BASEADDR_G(x) (((x) >> BASEADDR_S) & BASEADDR_M)
+
 #define TCAMINTPERR_S    13
 #define TCAMINTPERR_V(x) ((x) << TCAMINTPERR_S)
 #define TCAMINTPERR_F    TCAMINTPERR_V(1U)
@@ -2740,10 +2932,11 @@
 #define EDC_H_BIST_DATA_PATTERN_A	0x50010
 #define EDC_H_BIST_STATUS_RDATA_A	0x50028
 
+#define EDC_H_ECC_ERR_ADDR_A		0x50084
 #define EDC_T51_BASE_ADDR		0x50800
 
-#define EDC_STRIDE_T5 (EDC_T51_BASE_ADDR - EDC_T50_BASE_ADDR)
-#define EDC_REG_T5(reg, idx) (reg + EDC_STRIDE_T5 * idx)
+#define EDC_T5_STRIDE (EDC_T51_BASE_ADDR - EDC_T50_BASE_ADDR)
+#define EDC_T5_REG(reg, idx) (reg + EDC_T5_STRIDE * idx)
 
 #define PL_VF_REV_A 0x4
 #define PL_VF_WHOAMI_A 0x0
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h
index 32b213559b02..92bafa793de6 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_version.h
@@ -40,14 +40,25 @@
 #define T4FW_VERSION_MICRO 0x20
 #define T4FW_VERSION_BUILD 0x00
 
+#define T4FW_MIN_VERSION_MAJOR 0x01
+#define T4FW_MIN_VERSION_MINOR 0x04
+#define T4FW_MIN_VERSION_MICRO 0x00
+
 #define T5FW_VERSION_MAJOR 0x01
 #define T5FW_VERSION_MINOR 0x0D
 #define T5FW_VERSION_MICRO 0x20
 #define T5FW_VERSION_BUILD 0x00
 
+#define T5FW_MIN_VERSION_MAJOR 0x00
+#define T5FW_MIN_VERSION_MINOR 0x00
+#define T5FW_MIN_VERSION_MICRO 0x00
+
 #define T6FW_VERSION_MAJOR 0x01
 #define T6FW_VERSION_MINOR 0x0D
 #define T6FW_VERSION_MICRO 0x2D
 #define T6FW_VERSION_BUILD 0x00
 
+#define T6FW_MIN_VERSION_MAJOR 0x00
+#define T6FW_MIN_VERSION_MINOR 0x00
+#define T6FW_MIN_VERSION_MICRO 0x00
 #endif
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
index ad53e5ad2acd..fa3786a9d30e 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c
@@ -1898,7 +1898,10 @@ static int napi_rx_handler(struct napi_struct *napi, int budget)
 		rspq->unhandled_irqs++;
 
 	val = CIDXINC_V(work_done) | SEINTARM_V(intr_params);
-	if (is_t4(rspq->adapter->params.chip)) {
+	/* If we don't have access to the new User GTS (T5+), use the old
+	 * doorbell mechanism; otherwise use the new BAR2 mechanism.
+	 */
+	if (unlikely(!rspq->bar2_addr)) {
 		t4_write_reg(rspq->adapter,
 			     T4VF_SGE_BASE_ADDR + SGE_VF_GTS,
 			     val | INGRESSQID_V((u32)rspq->cntxt_id));
@@ -1998,10 +2001,13 @@ static unsigned int process_intrq(struct adapter *adapter)
 	}
 
 	val = CIDXINC_V(work_done) | SEINTARM_V(intrq->intr_params);
-	if (is_t4(adapter->params.chip))
+	/* If we don't have access to the new User GTS (T5+), use the old
+	 * doorbell mechanism; otherwise use the new BAR2 mechanism.
+	 */
+	if (unlikely(!intrq->bar2_addr)) {
 		t4_write_reg(adapter, T4VF_SGE_BASE_ADDR + SGE_VF_GTS,
 			     val | INGRESSQID_V(intrq->cntxt_id));
-	else {
+	} else {
 		writel(val | INGRESSQID_V(intrq->bar2_qid),
 		       intrq->bar2_addr + SGE_UDB_GTS);
 		wmb();
@@ -2662,8 +2668,22 @@ int t4vf_sge_init(struct adapter *adapter)
 	 * give it more Free List entries.  (Note that the SGE's Egress
 	 * Congestion Threshold is in units of 2 Free List pointers.)
 	 */
-	s->fl_starve_thres
-		= EGRTHRESHOLD_G(sge_params->sge_congestion_control)*2 + 1;
+	switch (CHELSIO_CHIP_VERSION(adapter->params.chip)) {
+	case CHELSIO_T4:
+		s->fl_starve_thres =
+		   EGRTHRESHOLD_G(sge_params->sge_congestion_control);
+		break;
+	case CHELSIO_T5:
+		s->fl_starve_thres =
+		   EGRTHRESHOLDPACKING_G(sge_params->sge_congestion_control);
+		break;
+	case CHELSIO_T6:
+	default:
+		s->fl_starve_thres =
+		   T6_EGRTHRESHOLDPACKING_G(sge_params->sge_congestion_control);
+		break;
+	}
+	s->fl_starve_thres = s->fl_starve_thres * 2 + 1;
 
 	/*
 	 * Set up tasklet timers.
diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
index 0db6dc9e9ed2..63dd5fdac5b9 100644
--- a/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb4vf/t4vf_hw.c
@@ -619,7 +619,8 @@ int t4vf_get_sge_params(struct adapter *adapter)
 		 */
 		whoami = t4_read_reg(adapter,
 				     T4VF_PL_BASE_ADDR + PL_VF_WHOAMI_A);
-		pf = SOURCEPF_G(whoami);
+		pf = CHELSIO_CHIP_VERSION(adapter->params.chip) <= CHELSIO_T5 ?
+			SOURCEPF_G(whoami) : T6_SOURCEPF_G(whoami);
 
 		s_hps = (HOSTPAGESIZEPF0_S +
 			 (HOSTPAGESIZEPF1_S - HOSTPAGESIZEPF0_S) * pf);