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

Pull networking changes from David Miller:

 1) GRE now works over ipv6, from Dmitry Kozlov.

 2) Make SCTP more network namespace aware, from Eric Biederman.

 3) TEAM driver now works with non-ethernet devices, from Jiri Pirko.

 4) Make openvswitch network namespace aware, from Pravin B Shelar.

 5) IPV6 NAT implementation, from Patrick McHardy.

 6) Server side support for TCP Fast Open, from Jerry Chu and others.

 7) Packet BPF filter supports MOD and XOR, from Eric Dumazet and Daniel
    Borkmann.

 8) Increate the loopback default MTU to 64K, from Eric Dumazet.

 9) Use a per-task rather than per-socket page fragment allocator for
    outgoing networking traffic.  This benefits processes that have very
    many mostly idle sockets, which is quite common.

    From Eric Dumazet.

10) Use up to 32K for page fragment allocations, with fallbacks to
    smaller sizes when higher order page allocations fail.  Benefits are
    a) less segments for driver to process b) less calls to page
    allocator c) less waste of space.

    From Eric Dumazet.

11) Allow GRO to be used on GRE tunnels, from Eric Dumazet.

12) VXLAN device driver, one way to handle VLAN issues such as the
    limitation of 4096 VLAN IDs yet still have some level of isolation.
    From Stephen Hemminger.

13) As usual there is a large boatload of driver changes, with the scale
    perhaps tilted towards the wireless side this time around.

Fix up various fairly trivial conflicts, mostly caused by the user
namespace changes.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1012 commits)
  hyperv: Add buffer for extended info after the RNDIS response message.
  hyperv: Report actual status in receive completion packet
  hyperv: Remove extra allocated space for recv_pkt_list elements
  hyperv: Fix page buffer handling in rndis_filter_send_request()
  hyperv: Fix the missing return value in rndis_filter_set_packet_filter()
  hyperv: Fix the max_xfer_size in RNDIS initialization
  vxlan: put UDP socket in correct namespace
  vxlan: Depend on CONFIG_INET
  sfc: Fix the reported priorities of different filter types
  sfc: Remove EFX_FILTER_FLAG_RX_OVERRIDE_IP
  sfc: Fix loopback self-test with separate_tx_channels=1
  sfc: Fix MCDI structure field lookup
  sfc: Add parentheses around use of bitfield macro arguments
  sfc: Fix null function pointer in efx_sriov_channel_type
  vxlan: virtual extensible lan
  igmp: export symbol ip_mc_leave_group
  netlink: add attributes to fdb interface
  tg3: unconditionally select HWMON support when tg3 is enabled.
  Revert "net: ti cpsw ethernet: allow reading phy interface mode from DT"
  gre: fix sparse warning
  ...

1 files changed, 67 insertions, 45 deletions

diff --git a/drivers/net/wireless/ti/wlcore/tx.c b/drivers/net/wireless/ti/wlcore/tx.c
index f0081f74648..a90d3cd0940 100644
--- a/drivers/net/wireless/ti/wlcore/tx.c
+++ b/drivers/net/wireless/ti/wlcore/tx.c
@@ -130,16 +130,13 @@ bool wl12xx_is_dummy_packet(struct wl1271 *wl, struct sk_buff *skb)
 }
 EXPORT_SYMBOL(wl12xx_is_dummy_packet);
 
-u8 wl12xx_tx_get_hlid_ap(struct wl1271 *wl, struct wl12xx_vif *wlvif,
-			 struct sk_buff *skb)
+static u8 wl12xx_tx_get_hlid_ap(struct wl1271 *wl, struct wl12xx_vif *wlvif,
+				struct sk_buff *skb, struct ieee80211_sta *sta)
 {
-	struct ieee80211_tx_info *control = IEEE80211_SKB_CB(skb);
-
-	if (control->control.sta) {
+	if (sta) {
 		struct wl1271_station *wl_sta;
 
-		wl_sta = (struct wl1271_station *)
-				control->control.sta->drv_priv;
+		wl_sta = (struct wl1271_station *)sta->drv_priv;
 		return wl_sta->hlid;
 	} else {
 		struct ieee80211_hdr *hdr;
@@ -156,7 +153,7 @@ u8 wl12xx_tx_get_hlid_ap(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 }
 
 u8 wl12xx_tx_get_hlid(struct wl1271 *wl, struct wl12xx_vif *wlvif,
-		      struct sk_buff *skb)
+		      struct sk_buff *skb, struct ieee80211_sta *sta)
 {
 	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
 
@@ -164,7 +161,7 @@ u8 wl12xx_tx_get_hlid(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 		return wl->system_hlid;
 
 	if (wlvif->bss_type == BSS_TYPE_AP_BSS)
-		return wl12xx_tx_get_hlid_ap(wl, wlvif, skb);
+		return wl12xx_tx_get_hlid_ap(wl, wlvif, skb, sta);
 
 	if ((test_bit(WLVIF_FLAG_STA_ASSOCIATED, &wlvif->flags) ||
 	     test_bit(WLVIF_FLAG_IBSS_JOINED, &wlvif->flags)) &&
@@ -196,7 +193,7 @@ static int wl1271_tx_allocate(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 	int id, ret = -EBUSY, ac;
 	u32 spare_blocks;
 
-	if (buf_offset + total_len > WL1271_AGGR_BUFFER_SIZE)
+	if (buf_offset + total_len > wl->aggr_buf_size)
 		return -EAGAIN;
 
 	spare_blocks = wlcore_hw_get_spare_blocks(wl, is_gem);
@@ -322,8 +319,12 @@ static void wl1271_tx_fill_hdr(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 		if (hlid == wlvif->ap.global_hlid)
 			rate_idx = wlvif->ap.mgmt_rate_idx;
 		else if (hlid == wlvif->ap.bcast_hlid ||
-			 skb->protocol == cpu_to_be16(ETH_P_PAE))
-			/* send AP bcast and EAPOLs using the min basic rate */
+			 skb->protocol == cpu_to_be16(ETH_P_PAE) ||
+			 !ieee80211_is_data(frame_control))
+			/*
+			 * send non-data, bcast and EAPOLs using the
+			 * min basic rate
+			 */
 			rate_idx = wlvif->ap.bcast_rate_idx;
 		else
 			rate_idx = wlvif->ap.ucast_rate_idx[ac];
@@ -344,13 +345,12 @@ static void wl1271_tx_fill_hdr(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 
 /* caller must hold wl->mutex */
 static int wl1271_prepare_tx_frame(struct wl1271 *wl, struct wl12xx_vif *wlvif,
-				   struct sk_buff *skb, u32 buf_offset)
+				   struct sk_buff *skb, u32 buf_offset, u8 hlid)
 {
 	struct ieee80211_tx_info *info;
 	u32 extra = 0;
 	int ret = 0;
 	u32 total_len;
-	u8 hlid;
 	bool is_dummy;
 	bool is_gem = false;
 
@@ -359,9 +359,13 @@ static int wl1271_prepare_tx_frame(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 		return -EINVAL;
 	}
 
+	if (hlid == WL12XX_INVALID_LINK_ID) {
+		wl1271_error("invalid hlid. dropping skb 0x%p", skb);
+		return -EINVAL;
+	}
+
 	info = IEEE80211_SKB_CB(skb);
 
-	/* TODO: handle dummy packets on multi-vifs */
 	is_dummy = wl12xx_is_dummy_packet(wl, skb);
 
 	if ((wl->quirks & WLCORE_QUIRK_TKIP_HEADER_SPACE) &&
@@ -386,11 +390,6 @@ static int wl1271_prepare_tx_frame(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 
 		is_gem = (cipher == WL1271_CIPHER_SUITE_GEM);
 	}
-	hlid = wl12xx_tx_get_hlid(wl, wlvif, skb);
-	if (hlid == WL12XX_INVALID_LINK_ID) {
-		wl1271_error("invalid hlid. dropping skb 0x%p", skb);
-		return -EINVAL;
-	}
 
 	ret = wl1271_tx_allocate(wl, wlvif, skb, extra, buf_offset, hlid,
 				 is_gem);
@@ -517,7 +516,8 @@ static struct sk_buff *wl12xx_lnk_skb_dequeue(struct wl1271 *wl,
 }
 
 static struct sk_buff *wl12xx_vif_skb_dequeue(struct wl1271 *wl,
-					      struct wl12xx_vif *wlvif)
+					      struct wl12xx_vif *wlvif,
+					      u8 *hlid)
 {
 	struct sk_buff *skb = NULL;
 	int i, h, start_hlid;
@@ -544,10 +544,11 @@ static struct sk_buff *wl12xx_vif_skb_dequeue(struct wl1271 *wl,
 	if (!skb)
 		wlvif->last_tx_hlid = 0;
 
+	*hlid = wlvif->last_tx_hlid;
 	return skb;
 }
 
-static struct sk_buff *wl1271_skb_dequeue(struct wl1271 *wl)
+static struct sk_buff *wl1271_skb_dequeue(struct wl1271 *wl, u8 *hlid)
 {
 	unsigned long flags;
 	struct wl12xx_vif *wlvif = wl->last_wlvif;
@@ -556,7 +557,7 @@ static struct sk_buff *wl1271_skb_dequeue(struct wl1271 *wl)
 	/* continue from last wlvif (round robin) */
 	if (wlvif) {
 		wl12xx_for_each_wlvif_continue(wl, wlvif) {
-			skb = wl12xx_vif_skb_dequeue(wl, wlvif);
+			skb = wl12xx_vif_skb_dequeue(wl, wlvif, hlid);
 			if (skb) {
 				wl->last_wlvif = wlvif;
 				break;
@@ -565,13 +566,15 @@ static struct sk_buff *wl1271_skb_dequeue(struct wl1271 *wl)
 	}
 
 	/* dequeue from the system HLID before the restarting wlvif list */
-	if (!skb)
+	if (!skb) {
 		skb = wl12xx_lnk_skb_dequeue(wl, &wl->links[wl->system_hlid]);
+		*hlid = wl->system_hlid;
+	}
 
 	/* do a new pass over the wlvif list */
 	if (!skb) {
 		wl12xx_for_each_wlvif(wl, wlvif) {
-			skb = wl12xx_vif_skb_dequeue(wl, wlvif);
+			skb = wl12xx_vif_skb_dequeue(wl, wlvif, hlid);
 			if (skb) {
 				wl->last_wlvif = wlvif;
 				break;
@@ -591,6 +594,7 @@ static struct sk_buff *wl1271_skb_dequeue(struct wl1271 *wl)
 		int q;
 
 		skb = wl->dummy_packet;
+		*hlid = wl->system_hlid;
 		q = wl1271_tx_get_queue(skb_get_queue_mapping(skb));
 		spin_lock_irqsave(&wl->wl_lock, flags);
 		WARN_ON_ONCE(wl->tx_queue_count[q] <= 0);
@@ -602,7 +606,7 @@ static struct sk_buff *wl1271_skb_dequeue(struct wl1271 *wl)
 }
 
 static void wl1271_skb_queue_head(struct wl1271 *wl, struct wl12xx_vif *wlvif,
-				  struct sk_buff *skb)
+				  struct sk_buff *skb, u8 hlid)
 {
 	unsigned long flags;
 	int q = wl1271_tx_get_queue(skb_get_queue_mapping(skb));
@@ -610,7 +614,6 @@ static void wl1271_skb_queue_head(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 	if (wl12xx_is_dummy_packet(wl, skb)) {
 		set_bit(WL1271_FLAG_DUMMY_PACKET_PENDING, &wl->flags);
 	} else {
-		u8 hlid = wl12xx_tx_get_hlid(wl, wlvif, skb);
 		skb_queue_head(&wl->links[hlid].tx_queue[q], skb);
 
 		/* make sure we dequeue the same packet next time */
@@ -686,26 +689,30 @@ int wlcore_tx_work_locked(struct wl1271 *wl)
 	unsigned long active_hlids[BITS_TO_LONGS(WL12XX_MAX_LINKS)] = {0};
 	int ret = 0;
 	int bus_ret = 0;
+	u8 hlid;
 
-	if (unlikely(wl->state == WL1271_STATE_OFF))
+	if (unlikely(wl->state != WLCORE_STATE_ON))
 		return 0;
 
-	while ((skb = wl1271_skb_dequeue(wl))) {
+	while ((skb = wl1271_skb_dequeue(wl, &hlid))) {
 		struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
 		bool has_data = false;
 
 		wlvif = NULL;
 		if (!wl12xx_is_dummy_packet(wl, skb) && info->control.vif)
 			wlvif = wl12xx_vif_to_data(info->control.vif);
+		else
+			hlid = wl->system_hlid;
 
 		has_data = wlvif && wl1271_tx_is_data_present(skb);
-		ret = wl1271_prepare_tx_frame(wl, wlvif, skb, buf_offset);
+		ret = wl1271_prepare_tx_frame(wl, wlvif, skb, buf_offset,
+					      hlid);
 		if (ret == -EAGAIN) {
 			/*
 			 * Aggregation buffer is full.
 			 * Flush buffer and try again.
 			 */
-			wl1271_skb_queue_head(wl, wlvif, skb);
+			wl1271_skb_queue_head(wl, wlvif, skb, hlid);
 
 			buf_offset = wlcore_hw_pre_pkt_send(wl, buf_offset,
 							    last_len);
@@ -722,7 +729,7 @@ int wlcore_tx_work_locked(struct wl1271 *wl)
 			 * Firmware buffer is full.
 			 * Queue back last skb, and stop aggregating.
 			 */
-			wl1271_skb_queue_head(wl, wlvif, skb);
+			wl1271_skb_queue_head(wl, wlvif, skb, hlid);
 			/* No work left, avoid scheduling redundant tx work */
 			set_bit(WL1271_FLAG_FW_TX_BUSY, &wl->flags);
 			goto out_ack;
@@ -732,7 +739,7 @@ int wlcore_tx_work_locked(struct wl1271 *wl)
 				 * fw still expects dummy packet,
 				 * so re-enqueue it
 				 */
-				wl1271_skb_queue_head(wl, wlvif, skb);
+				wl1271_skb_queue_head(wl, wlvif, skb, hlid);
 			else
 				ieee80211_free_txskb(wl->hw, skb);
 			goto out_ack;
@@ -1069,39 +1076,54 @@ void wl12xx_tx_reset(struct wl1271 *wl)
 /* caller must *NOT* hold wl->mutex */
 void wl1271_tx_flush(struct wl1271 *wl)
 {
-	unsigned long timeout;
+	unsigned long timeout, start_time;
 	int i;
-	timeout = jiffies + usecs_to_jiffies(WL1271_TX_FLUSH_TIMEOUT);
+	start_time = jiffies;
+	timeout = start_time + usecs_to_jiffies(WL1271_TX_FLUSH_TIMEOUT);
 
 	/* only one flush should be in progress, for consistent queue state */
 	mutex_lock(&wl->flush_mutex);
 
+	mutex_lock(&wl->mutex);
+	if (wl->tx_frames_cnt == 0 && wl1271_tx_total_queue_count(wl) == 0) {
+		mutex_unlock(&wl->mutex);
+		goto out;
+	}
+
 	wlcore_stop_queues(wl, WLCORE_QUEUE_STOP_REASON_FLUSH);
 
 	while (!time_after(jiffies, timeout)) {
-		mutex_lock(&wl->mutex);
-		wl1271_debug(DEBUG_TX, "flushing tx buffer: %d %d",
+		wl1271_debug(DEBUG_MAC80211, "flushing tx buffer: %d %d",
 			     wl->tx_frames_cnt,
 			     wl1271_tx_total_queue_count(wl));
+
+		/* force Tx and give the driver some time to flush data */
+		mutex_unlock(&wl->mutex);
+		if (wl1271_tx_total_queue_count(wl))
+			wl1271_tx_work(&wl->tx_work);
+		msleep(20);
+		mutex_lock(&wl->mutex);
+
 		if ((wl->tx_frames_cnt == 0) &&
 		    (wl1271_tx_total_queue_count(wl) == 0)) {
-			mutex_unlock(&wl->mutex);
-			goto out;
+			wl1271_debug(DEBUG_MAC80211, "tx flush took %d ms",
+				     jiffies_to_msecs(jiffies - start_time));
+			goto out_wake;
 		}
-		mutex_unlock(&wl->mutex);
-		msleep(1);
 	}
 
-	wl1271_warning("Unable to flush all TX buffers, timed out.");
+	wl1271_warning("Unable to flush all TX buffers, "
+		       "timed out (timeout %d ms",
+		       WL1271_TX_FLUSH_TIMEOUT / 1000);
 
 	/* forcibly flush all Tx buffers on our queues */
-	mutex_lock(&wl->mutex);
 	for (i = 0; i < WL12XX_MAX_LINKS; i++)
 		wl1271_tx_reset_link_queues(wl, i);
-	mutex_unlock(&wl->mutex);
 
-out:
+out_wake:
 	wlcore_wake_queues(wl, WLCORE_QUEUE_STOP_REASON_FLUSH);
+	mutex_unlock(&wl->mutex);
+out:
 	mutex_unlock(&wl->flush_mutex);
 }
 EXPORT_SYMBOL_GPL(wl1271_tx_flush);