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
  ...

1 files changed, 152 insertions, 8 deletions

diff --git a/drivers/net/wireless/iwlwifi/mvm/rs.c b/drivers/net/wireless/iwlwifi/mvm/rs.c
index daff1d0a8e4a..5ae9c8aa868f 100644
--- a/drivers/net/wireless/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/iwlwifi/mvm/rs.c
@@ -177,7 +177,8 @@ static bool rs_mimo_allow(struct iwl_mvm *mvm, struct ieee80211_sta *sta,
 
 	mvmsta = iwl_mvm_sta_from_mac80211(sta);
 	mvmvif = iwl_mvm_vif_from_mac80211(mvmsta->vif);
-	if (iwl_mvm_vif_low_latency(mvmvif) && mvmsta->vif->p2p)
+	if (IWL_MVM_RS_DISABLE_P2P_MIMO &&
+	    iwl_mvm_vif_low_latency(mvmvif) && mvmsta->vif->p2p)
 		return false;
 
 	if (mvm->nvm_data->sku_cap_mimo_disabled)
@@ -2403,7 +2404,7 @@ struct rs_init_rate_info {
 	u8 rate_idx;
 };
 
-static const struct rs_init_rate_info rs_init_rates_24ghz[] = {
+static const struct rs_init_rate_info rs_optimal_rates_24ghz_legacy[] = {
 	{ -60, IWL_RATE_54M_INDEX },
 	{ -64, IWL_RATE_48M_INDEX },
 	{ -68, IWL_RATE_36M_INDEX },
@@ -2416,7 +2417,7 @@ static const struct rs_init_rate_info rs_init_rates_24ghz[] = {
 	{ S8_MIN, IWL_RATE_1M_INDEX },
 };
 
-static const struct rs_init_rate_info rs_init_rates_5ghz[] = {
+static const struct rs_init_rate_info rs_optimal_rates_5ghz_legacy[] = {
 	{ -60, IWL_RATE_54M_INDEX },
 	{ -64, IWL_RATE_48M_INDEX },
 	{ -72, IWL_RATE_36M_INDEX },
@@ -2427,6 +2428,124 @@ static const struct rs_init_rate_info rs_init_rates_5ghz[] = {
 	{ S8_MIN, IWL_RATE_6M_INDEX },
 };
 
+static const struct rs_init_rate_info rs_optimal_rates_ht[] = {
+	{ -60, IWL_RATE_MCS_7_INDEX },
+	{ -64, IWL_RATE_MCS_6_INDEX },
+	{ -68, IWL_RATE_MCS_5_INDEX },
+	{ -72, IWL_RATE_MCS_4_INDEX },
+	{ -80, IWL_RATE_MCS_3_INDEX },
+	{ -84, IWL_RATE_MCS_2_INDEX },
+	{ -85, IWL_RATE_MCS_1_INDEX },
+	{ S8_MIN, IWL_RATE_MCS_0_INDEX},
+};
+
+static const struct rs_init_rate_info rs_optimal_rates_vht_20mhz[] = {
+	{ -60, IWL_RATE_MCS_8_INDEX },
+	{ -64, IWL_RATE_MCS_7_INDEX },
+	{ -68, IWL_RATE_MCS_6_INDEX },
+	{ -72, IWL_RATE_MCS_5_INDEX },
+	{ -80, IWL_RATE_MCS_4_INDEX },
+	{ -84, IWL_RATE_MCS_3_INDEX },
+	{ -85, IWL_RATE_MCS_2_INDEX },
+	{ -87, IWL_RATE_MCS_1_INDEX },
+	{ S8_MIN, IWL_RATE_MCS_0_INDEX},
+};
+
+static const struct rs_init_rate_info rs_optimal_rates_vht_40_80mhz[] = {
+	{ -60, IWL_RATE_MCS_9_INDEX },
+	{ -64, IWL_RATE_MCS_8_INDEX },
+	{ -68, IWL_RATE_MCS_7_INDEX },
+	{ -72, IWL_RATE_MCS_6_INDEX },
+	{ -80, IWL_RATE_MCS_5_INDEX },
+	{ -84, IWL_RATE_MCS_4_INDEX },
+	{ -85, IWL_RATE_MCS_3_INDEX },
+	{ -87, IWL_RATE_MCS_2_INDEX },
+	{ -88, IWL_RATE_MCS_1_INDEX },
+	{ S8_MIN, IWL_RATE_MCS_0_INDEX },
+};
+
+/* Init the optimal rate based on STA caps
+ * This combined with rssi is used to report the last tx rate
+ * to userspace when we haven't transmitted enough frames.
+ */
+static void rs_init_optimal_rate(struct iwl_mvm *mvm,
+				 struct ieee80211_sta *sta,
+				 struct iwl_lq_sta *lq_sta)
+{
+	struct rs_rate *rate = &lq_sta->optimal_rate;
+
+	if (lq_sta->max_mimo2_rate_idx != IWL_RATE_INVALID)
+		rate->type = lq_sta->is_vht ? LQ_VHT_MIMO2 : LQ_HT_MIMO2;
+	else if (lq_sta->max_siso_rate_idx != IWL_RATE_INVALID)
+		rate->type = lq_sta->is_vht ? LQ_VHT_SISO : LQ_HT_SISO;
+	else if (lq_sta->band == IEEE80211_BAND_5GHZ)
+		rate->type = LQ_LEGACY_A;
+	else
+		rate->type = LQ_LEGACY_G;
+
+	rate->bw = rs_bw_from_sta_bw(sta);
+	rate->sgi = rs_sgi_allow(mvm, sta, rate, NULL);
+
+	/* ANT/LDPC/STBC aren't relevant for the rate reported to userspace */
+
+	if (is_mimo(rate)) {
+		lq_sta->optimal_rate_mask = lq_sta->active_mimo2_rate;
+	} else if (is_siso(rate)) {
+		lq_sta->optimal_rate_mask = lq_sta->active_siso_rate;
+	} else {
+		lq_sta->optimal_rate_mask = lq_sta->active_legacy_rate;
+
+		if (lq_sta->band == IEEE80211_BAND_5GHZ) {
+			lq_sta->optimal_rates = rs_optimal_rates_5ghz_legacy;
+			lq_sta->optimal_nentries =
+				ARRAY_SIZE(rs_optimal_rates_5ghz_legacy);
+		} else {
+			lq_sta->optimal_rates = rs_optimal_rates_24ghz_legacy;
+			lq_sta->optimal_nentries =
+				ARRAY_SIZE(rs_optimal_rates_24ghz_legacy);
+		}
+	}
+
+	if (is_vht(rate)) {
+		if (rate->bw == RATE_MCS_CHAN_WIDTH_20) {
+			lq_sta->optimal_rates = rs_optimal_rates_vht_20mhz;
+			lq_sta->optimal_nentries =
+				ARRAY_SIZE(rs_optimal_rates_vht_20mhz);
+		} else {
+			lq_sta->optimal_rates = rs_optimal_rates_vht_40_80mhz;
+			lq_sta->optimal_nentries =
+				ARRAY_SIZE(rs_optimal_rates_vht_40_80mhz);
+		}
+	} else if (is_ht(rate)) {
+		lq_sta->optimal_rates = rs_optimal_rates_ht;
+		lq_sta->optimal_nentries = ARRAY_SIZE(rs_optimal_rates_ht);
+	}
+}
+
+/* Compute the optimal rate index based on RSSI */
+static struct rs_rate *rs_get_optimal_rate(struct iwl_mvm *mvm,
+					   struct iwl_lq_sta *lq_sta)
+{
+	struct rs_rate *rate = &lq_sta->optimal_rate;
+	int i;
+
+	rate->index = find_first_bit(&lq_sta->optimal_rate_mask,
+				     BITS_PER_LONG);
+
+	for (i = 0; i < lq_sta->optimal_nentries; i++) {
+		int rate_idx = lq_sta->optimal_rates[i].rate_idx;
+
+		if ((lq_sta->pers.last_rssi >= lq_sta->optimal_rates[i].rssi) &&
+		    (BIT(rate_idx) & lq_sta->optimal_rate_mask)) {
+			rate->index = rate_idx;
+			break;
+		}
+	}
+
+	rs_dump_rate(mvm, rate, "OPTIMAL RATE");
+	return rate;
+}
+
 /* Choose an initial legacy rate and antenna to use based on the RSSI
  * of last Rx
  */
@@ -2468,12 +2587,12 @@ static void rs_get_initial_rate(struct iwl_mvm *mvm,
 
 	if (band == IEEE80211_BAND_5GHZ) {
 		rate->type = LQ_LEGACY_A;
-		initial_rates = rs_init_rates_5ghz;
-		nentries = ARRAY_SIZE(rs_init_rates_5ghz);
+		initial_rates = rs_optimal_rates_5ghz_legacy;
+		nentries = ARRAY_SIZE(rs_optimal_rates_5ghz_legacy);
 	} else {
 		rate->type = LQ_LEGACY_G;
-		initial_rates = rs_init_rates_24ghz;
-		nentries = ARRAY_SIZE(rs_init_rates_24ghz);
+		initial_rates = rs_optimal_rates_24ghz_legacy;
+		nentries = ARRAY_SIZE(rs_optimal_rates_24ghz_legacy);
 	}
 
 	if (IWL_MVM_RS_RSSI_BASED_INIT_RATE) {
@@ -2496,10 +2615,21 @@ void rs_update_last_rssi(struct iwl_mvm *mvm,
 			 struct iwl_lq_sta *lq_sta,
 			 struct ieee80211_rx_status *rx_status)
 {
+	int i;
+
 	lq_sta->pers.chains = rx_status->chains;
 	lq_sta->pers.chain_signal[0] = rx_status->chain_signal[0];
 	lq_sta->pers.chain_signal[1] = rx_status->chain_signal[1];
 	lq_sta->pers.chain_signal[2] = rx_status->chain_signal[2];
+	lq_sta->pers.last_rssi = S8_MIN;
+
+	for (i = 0; i < ARRAY_SIZE(lq_sta->pers.chain_signal); i++) {
+		if (!(lq_sta->pers.chains & BIT(i)))
+			continue;
+
+		if (lq_sta->pers.chain_signal[i] > lq_sta->pers.last_rssi)
+			lq_sta->pers.last_rssi = lq_sta->pers.chain_signal[i];
+	}
 }
 
 /**
@@ -2538,6 +2668,7 @@ static void rs_initialize_lq(struct iwl_mvm *mvm,
 	rate = &tbl->rate;
 
 	rs_get_initial_rate(mvm, lq_sta, band, rate);
+	rs_init_optimal_rate(mvm, sta, lq_sta);
 
 	WARN_ON_ONCE(rate->ant != ANT_A && rate->ant != ANT_B);
 	if (rate->ant == ANT_A)
@@ -2560,6 +2691,8 @@ static void rs_get_rate(void *mvm_r, struct ieee80211_sta *sta, void *mvm_sta,
 	struct iwl_mvm *mvm __maybe_unused = IWL_OP_MODE_GET_MVM(op_mode);
 	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 	struct iwl_lq_sta *lq_sta = mvm_sta;
+	struct rs_rate *optimal_rate;
+	u32 last_ucode_rate;
 
 	if (sta && !iwl_mvm_sta_from_mac80211(sta)->vif) {
 		/* if vif isn't initialized mvm doesn't know about
@@ -2583,8 +2716,18 @@ static void rs_get_rate(void *mvm_r, struct ieee80211_sta *sta, void *mvm_sta,
 
 	iwl_mvm_hwrate_to_tx_rate(lq_sta->last_rate_n_flags,
 				  info->band, &info->control.rates[0]);
-
 	info->control.rates[0].count = 1;
+
+	/* Report the optimal rate based on rssi and STA caps if we haven't
+	 * converged yet (too little traffic) or exploring other modulations
+	 */
+	if (lq_sta->rs_state != RS_STATE_STAY_IN_COLUMN) {
+		optimal_rate = rs_get_optimal_rate(mvm, lq_sta);
+		last_ucode_rate = ucode_rate_from_rs_rate(mvm,
+							  optimal_rate);
+		iwl_mvm_hwrate_to_tx_rate(last_ucode_rate, info->band,
+					  &txrc->reported_rate);
+	}
 }
 
 static void *rs_alloc_sta(void *mvm_rate, struct ieee80211_sta *sta,
@@ -2605,6 +2748,7 @@ static void *rs_alloc_sta(void *mvm_rate, struct ieee80211_sta *sta,
 #endif
 	lq_sta->pers.chains = 0;
 	memset(lq_sta->pers.chain_signal, 0, sizeof(lq_sta->pers.chain_signal));
+	lq_sta->pers.last_rssi = S8_MIN;
 
 	return &sta_priv->lq_sta;
 }