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

Pull networking updates from David Miller:

 1) More iov_iter conversion work from Al Viro.

    [ The "crypto: switch af_alg_make_sg() to iov_iter" commit was
      wrong, and this pull actually adds an extra commit on top of the
      branch I'm pulling to fix that up, so that the pre-merge state is
      ok.   - Linus ]

 2) Various optimizations to the ipv4 forwarding information base trie
    lookup implementation.  From Alexander Duyck.

 3) Remove sock_iocb altogether, from CHristoph Hellwig.

 4) Allow congestion control algorithm selection via routing metrics.
    From Daniel Borkmann.

 5) Make ipv4 uncached route list per-cpu, from Eric Dumazet.

 6) Handle rfs hash collisions more gracefully, also from Eric Dumazet.

 7) Add xmit_more support to r8169, e1000, and e1000e drivers.  From
    Florian Westphal.

 8) Transparent Ethernet Bridging support for GRO, from Jesse Gross.

 9) Add BPF packet actions to packet scheduler, from Jiri Pirko.

10) Add support for uniqu flow IDs to openvswitch, from Joe Stringer.

11) New NetCP ethernet driver, from Muralidharan Karicheri and Wingman
    Kwok.

12) More sanely handle out-of-window dupacks, which can result in
    serious ACK storms.  From Neal Cardwell.

13) Various rhashtable bug fixes and enhancements, from Herbert Xu,
    Patrick McHardy, and Thomas Graf.

14) Support xmit_more in be2net, from Sathya Perla.

15) Group Policy extensions for vxlan, from Thomas Graf.

16) Remove Checksum Offload support for vxlan, from Tom Herbert.

17) Like ipv4, support lockless transmit over ipv6 UDP sockets.  From
    Vlad Yasevich.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1494+1 commits)
  crypto: fix af_alg_make_sg() conversion to iov_iter
  ipv4: Namespecify TCP PMTU mechanism
  i40e: Fix for stats init function call in Rx setup
  tcp: don't include Fast Open option in SYN-ACK on pure SYN-data
  openvswitch: Only set TUNNEL_VXLAN_OPT if VXLAN-GBP metadata is set
  ipv6: Make __ipv6_select_ident static
  ipv6: Fix fragment id assignment on LE arches.
  bridge: Fix inability to add non-vlan fdb entry
  net: Mellanox: Delete unnecessary checks before the function call "vunmap"
  cxgb4: Add support in cxgb4 to get expansion rom version via ethtool
  ethtool: rename reserved1 memeber in ethtool_drvinfo for expansion ROM version
  net: dsa: Remove redundant phy_attach()
  IB/mlx4: Reset flow support for IB kernel ULPs
  IB/mlx4: Always use the correct port for mirrored multicast attachments
  net/bonding: Fix potential bad memory access during bonding events
  tipc: remove tipc_snprintf
  tipc: nl compat add noop and remove legacy nl framework
  tipc: convert legacy nl stats show to nl compat
  tipc: convert legacy nl net id get to nl compat
  tipc: convert legacy nl net id set to nl compat
  ...

1 files changed, 36 insertions, 18 deletions

diff --git a/drivers/net/wireless/brcm80211/brcmfmac/msgbuf.c b/drivers/net/wireless/brcm80211/brcmfmac/msgbuf.c
index 456944a6a2db..6262612dec45 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/msgbuf.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/msgbuf.c
@@ -73,6 +73,8 @@
 #define BRCMF_MSGBUF_TX_FLUSH_CNT1		32
 #define BRCMF_MSGBUF_TX_FLUSH_CNT2		96
 
+#define BRCMF_MSGBUF_DELAY_TXWORKER_THRS	64
+#define BRCMF_MSGBUF_TRICKLE_TXWORKER_THRS	32
 
 struct msgbuf_common_hdr {
 	u8				msgtype;
@@ -583,7 +585,7 @@ brcmf_msgbuf_flowring_create_worker(struct brcmf_msgbuf *msgbuf,
 	u32 flowid;
 	void *dma_buf;
 	u32 dma_sz;
-	long long address;
+	u64 address;
 	int err;
 
 	flowid = work->flowid;
@@ -620,7 +622,7 @@ brcmf_msgbuf_flowring_create_worker(struct brcmf_msgbuf *msgbuf,
 					   BRCMF_NROF_H2D_COMMON_MSGRINGS);
 	memcpy(create->sa, work->sa, ETH_ALEN);
 	memcpy(create->da, work->da, ETH_ALEN);
-	address = (long long)(long)msgbuf->flowring_dma_handle[flowid];
+	address = (u64)msgbuf->flowring_dma_handle[flowid];
 	create->flow_ring_addr.high_addr = cpu_to_le32(address >> 32);
 	create->flow_ring_addr.low_addr = cpu_to_le32(address & 0xffffffff);
 	create->max_items = cpu_to_le16(BRCMF_H2D_TXFLOWRING_MAX_ITEM);
@@ -698,7 +700,7 @@ static void brcmf_msgbuf_txflow(struct brcmf_msgbuf *msgbuf, u8 flowid)
 	dma_addr_t physaddr;
 	u32 pktid;
 	struct msgbuf_tx_msghdr *tx_msghdr;
-	long long address;
+	u64 address;
 
 	commonring = msgbuf->flowrings[flowid];
 	if (!brcmf_commonring_write_available(commonring))
@@ -742,13 +744,14 @@ static void brcmf_msgbuf_txflow(struct brcmf_msgbuf *msgbuf, u8 flowid)
 		tx_msghdr->seg_cnt = 1;
 		memcpy(tx_msghdr->txhdr, skb->data, ETH_HLEN);
 		tx_msghdr->data_len = cpu_to_le16(skb->len - ETH_HLEN);
-		address = (long long)(long)physaddr;
+		address = (u64)physaddr;
 		tx_msghdr->data_buf_addr.high_addr = cpu_to_le32(address >> 32);
 		tx_msghdr->data_buf_addr.low_addr =
 			cpu_to_le32(address & 0xffffffff);
 		tx_msghdr->metadata_buf_len = 0;
 		tx_msghdr->metadata_buf_addr.high_addr = 0;
 		tx_msghdr->metadata_buf_addr.low_addr = 0;
+		atomic_inc(&commonring->outstanding_tx);
 		if (count >= BRCMF_MSGBUF_TX_FLUSH_CNT2) {
 			brcmf_commonring_write_complete(commonring);
 			count = 0;
@@ -773,10 +776,16 @@ static void brcmf_msgbuf_txflow_worker(struct work_struct *worker)
 }
 
 
-static int brcmf_msgbuf_schedule_txdata(struct brcmf_msgbuf *msgbuf, u32 flowid)
+static int brcmf_msgbuf_schedule_txdata(struct brcmf_msgbuf *msgbuf, u32 flowid,
+					bool force)
 {
+	struct brcmf_commonring *commonring;
+
 	set_bit(flowid, msgbuf->flow_map);
-	queue_work(msgbuf->txflow_wq, &msgbuf->txflow_work);
+	commonring = msgbuf->flowrings[flowid];
+	if ((force) || (atomic_read(&commonring->outstanding_tx) <
+			BRCMF_MSGBUF_DELAY_TXWORKER_THRS))
+		queue_work(msgbuf->txflow_wq, &msgbuf->txflow_work);
 
 	return 0;
 }
@@ -797,7 +806,7 @@ static int brcmf_msgbuf_txdata(struct brcmf_pub *drvr, int ifidx,
 			return -ENOMEM;
 	}
 	brcmf_flowring_enqueue(flow, flowid, skb);
-	brcmf_msgbuf_schedule_txdata(msgbuf, flowid);
+	brcmf_msgbuf_schedule_txdata(msgbuf, flowid, false);
 
 	return 0;
 }
@@ -854,6 +863,7 @@ brcmf_msgbuf_process_ioctl_complete(struct brcmf_msgbuf *msgbuf, void *buf)
 static void
 brcmf_msgbuf_process_txstatus(struct brcmf_msgbuf *msgbuf, void *buf)
 {
+	struct brcmf_commonring *commonring;
 	struct msgbuf_tx_status *tx_status;
 	u32 idx;
 	struct sk_buff *skb;
@@ -871,6 +881,8 @@ brcmf_msgbuf_process_txstatus(struct brcmf_msgbuf *msgbuf, void *buf)
 	}
 
 	set_bit(flowid, msgbuf->txstatus_done_map);
+	commonring = msgbuf->flowrings[flowid];
+	atomic_dec(&commonring->outstanding_tx);
 
 	brcmf_txfinalize(msgbuf->drvr, skb, tx_status->msg.ifidx, true);
 }
@@ -885,7 +897,7 @@ static u32 brcmf_msgbuf_rxbuf_data_post(struct brcmf_msgbuf *msgbuf, u32 count)
 	u32 pktlen;
 	dma_addr_t physaddr;
 	struct msgbuf_rx_bufpost *rx_bufpost;
-	long long address;
+	u64 address;
 	u32 pktid;
 	u32 i;
 
@@ -894,7 +906,7 @@ static u32 brcmf_msgbuf_rxbuf_data_post(struct brcmf_msgbuf *msgbuf, u32 count)
 							      count,
 							      &alloced);
 	if (!ret_ptr) {
-		brcmf_err("Failed to reserve space in commonring\n");
+		brcmf_dbg(MSGBUF, "Failed to reserve space in commonring\n");
 		return 0;
 	}
 
@@ -921,7 +933,7 @@ static u32 brcmf_msgbuf_rxbuf_data_post(struct brcmf_msgbuf *msgbuf, u32 count)
 		}
 
 		if (msgbuf->rx_metadata_offset) {
-			address = (long long)(long)physaddr;
+			address = (u64)physaddr;
 			rx_bufpost->metadata_buf_len =
 				cpu_to_le16(msgbuf->rx_metadata_offset);
 			rx_bufpost->metadata_buf_addr.high_addr =
@@ -936,7 +948,7 @@ static u32 brcmf_msgbuf_rxbuf_data_post(struct brcmf_msgbuf *msgbuf, u32 count)
 		rx_bufpost->msg.msgtype = MSGBUF_TYPE_RXBUF_POST;
 		rx_bufpost->msg.request_id = cpu_to_le32(pktid);
 
-		address = (long long)(long)physaddr;
+		address = (u64)physaddr;
 		rx_bufpost->data_buf_len = cpu_to_le16((u16)pktlen);
 		rx_bufpost->data_buf_addr.high_addr =
 			cpu_to_le32(address >> 32);
@@ -992,7 +1004,7 @@ brcmf_msgbuf_rxbuf_ctrl_post(struct brcmf_msgbuf *msgbuf, bool event_buf,
 	u32 pktlen;
 	dma_addr_t physaddr;
 	struct msgbuf_rx_ioctl_resp_or_event *rx_bufpost;
-	long long address;
+	u64 address;
 	u32 pktid;
 	u32 i;
 
@@ -1035,7 +1047,7 @@ brcmf_msgbuf_rxbuf_ctrl_post(struct brcmf_msgbuf *msgbuf, bool event_buf,
 				MSGBUF_TYPE_IOCTLRESP_BUF_POST;
 		rx_bufpost->msg.request_id = cpu_to_le32(pktid);
 
-		address = (long long)(long)physaddr;
+		address = (u64)physaddr;
 		rx_bufpost->host_buf_len = cpu_to_le16((u16)pktlen);
 		rx_bufpost->host_buf_addr.high_addr =
 			cpu_to_le32(address >> 32);
@@ -1181,7 +1193,7 @@ brcmf_msgbuf_process_flow_ring_create_response(struct brcmf_msgbuf *msgbuf,
 
 	brcmf_flowring_open(msgbuf->flow, flowid);
 
-	brcmf_msgbuf_schedule_txdata(msgbuf, flowid);
+	brcmf_msgbuf_schedule_txdata(msgbuf, flowid, true);
 }
 
 
@@ -1280,8 +1292,10 @@ int brcmf_proto_msgbuf_rx_trigger(struct device *dev)
 	struct brcmf_bus *bus_if = dev_get_drvdata(dev);
 	struct brcmf_pub *drvr = bus_if->drvr;
 	struct brcmf_msgbuf *msgbuf = (struct brcmf_msgbuf *)drvr->proto->pd;
+	struct brcmf_commonring *commonring;
 	void *buf;
 	u32 flowid;
+	int qlen;
 
 	buf = msgbuf->commonrings[BRCMF_D2H_MSGRING_RX_COMPLETE];
 	brcmf_msgbuf_process_rx(msgbuf, buf);
@@ -1293,8 +1307,12 @@ int brcmf_proto_msgbuf_rx_trigger(struct device *dev)
 	for_each_set_bit(flowid, msgbuf->txstatus_done_map,
 			 msgbuf->nrof_flowrings) {
 		clear_bit(flowid, msgbuf->txstatus_done_map);
-		if (brcmf_flowring_qlen(msgbuf->flow, flowid))
-			brcmf_msgbuf_schedule_txdata(msgbuf, flowid);
+		commonring = msgbuf->flowrings[flowid];
+		qlen = brcmf_flowring_qlen(msgbuf->flow, flowid);
+		if ((qlen > BRCMF_MSGBUF_TRICKLE_TXWORKER_THRS) ||
+		    ((qlen) && (atomic_read(&commonring->outstanding_tx) <
+				BRCMF_MSGBUF_TRICKLE_TXWORKER_THRS)))
+			brcmf_msgbuf_schedule_txdata(msgbuf, flowid, true);
 	}
 
 	return 0;
@@ -1348,7 +1366,7 @@ int brcmf_proto_msgbuf_attach(struct brcmf_pub *drvr)
 {
 	struct brcmf_bus_msgbuf *if_msgbuf;
 	struct brcmf_msgbuf *msgbuf;
-	long long address;
+	u64 address;
 	u32 count;
 
 	if_msgbuf = drvr->bus_if->msgbuf;
@@ -1379,7 +1397,7 @@ int brcmf_proto_msgbuf_attach(struct brcmf_pub *drvr)
 					     GFP_KERNEL);
 	if (!msgbuf->ioctbuf)
 		goto fail;
-	address = (long long)(long)msgbuf->ioctbuf_handle;
+	address = (u64)msgbuf->ioctbuf_handle;
 	msgbuf->ioctbuf_phys_hi = address >> 32;
 	msgbuf->ioctbuf_phys_lo = address & 0xffffffff;