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

Pull networking updates from David Miller:
 "Reasonably busy this cycle, but perhaps not as busy as in the 4.12
  merge window:

   1) Several optimizations for UDP processing under high load from
      Paolo Abeni.

   2) Support pacing internally in TCP when using the sch_fq packet
      scheduler for this is not practical. From Eric Dumazet.

   3) Support mutliple filter chains per qdisc, from Jiri Pirko.

   4) Move to 1ms TCP timestamp clock, from Eric Dumazet.

   5) Add batch dequeueing to vhost_net, from Jason Wang.

   6) Flesh out more completely SCTP checksum offload support, from
      Davide Caratti.

   7) More plumbing of extended netlink ACKs, from David Ahern, Pablo
      Neira Ayuso, and Matthias Schiffer.

   8) Add devlink support to nfp driver, from Simon Horman.

   9) Add RTM_F_FIB_MATCH flag to RTM_GETROUTE queries, from Roopa
      Prabhu.

  10) Add stack depth tracking to BPF verifier and use this information
      in the various eBPF JITs. From Alexei Starovoitov.

  11) Support XDP on qed device VFs, from Yuval Mintz.

  12) Introduce BPF PROG ID for better introspection of installed BPF
      programs. From Martin KaFai Lau.

  13) Add bpf_set_hash helper for TC bpf programs, from Daniel Borkmann.

  14) For loads, allow narrower accesses in bpf verifier checking, from
      Yonghong Song.

  15) Support MIPS in the BPF selftests and samples infrastructure, the
      MIPS eBPF JIT will be merged in via the MIPS GIT tree. From David
      Daney.

  16) Support kernel based TLS, from Dave Watson and others.

  17) Remove completely DST garbage collection, from Wei Wang.

  18) Allow installing TCP MD5 rules using prefixes, from Ivan
      Delalande.

  19) Add XDP support to Intel i40e driver, from Björn Töpel

  20) Add support for TC flower offload in nfp driver, from Simon
      Horman, Pieter Jansen van Vuuren, Benjamin LaHaise, Jakub
      Kicinski, and Bert van Leeuwen.

  21) IPSEC offloading support in mlx5, from Ilan Tayari.

  22) Add HW PTP support to macb driver, from Rafal Ozieblo.

  23) Networking refcount_t conversions, From Elena Reshetova.

  24) Add sock_ops support to BPF, from Lawrence Brako. This is useful
      for tuning the TCP sockopt settings of a group of applications,
      currently via CGROUPs"

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1899 commits)
  net: phy: dp83867: add workaround for incorrect RX_CTRL pin strap
  dt-bindings: phy: dp83867: provide a workaround for incorrect RX_CTRL pin strap
  cxgb4: Support for get_ts_info ethtool method
  cxgb4: Add PTP Hardware Clock (PHC) support
  cxgb4: time stamping interface for PTP
  nfp: default to chained metadata prepend format
  nfp: remove legacy MAC address lookup
  nfp: improve order of interfaces in breakout mode
  net: macb: remove extraneous return when MACB_EXT_DESC is defined
  bpf: add missing break in for the TCP_BPF_SNDCWND_CLAMP case
  bpf: fix return in load_bpf_file
  mpls: fix rtm policy in mpls_getroute
  net, ax25: convert ax25_cb.refcount from atomic_t to refcount_t
  net, ax25: convert ax25_route.refcount from atomic_t to refcount_t
  net, ax25: convert ax25_uid_assoc.refcount from atomic_t to refcount_t
  net, sctp: convert sctp_ep_common.refcnt from atomic_t to refcount_t
  net, sctp: convert sctp_transport.refcnt from atomic_t to refcount_t
  net, sctp: convert sctp_chunk.refcnt from atomic_t to refcount_t
  net, sctp: convert sctp_datamsg.refcnt from atomic_t to refcount_t
  net, sctp: convert sctp_auth_bytes.refcnt from atomic_t to refcount_t
  ...

1 files changed, 173 insertions, 11 deletions

diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
index 9c9bfbb..a3795ba 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c
@@ -49,6 +49,7 @@
  *
  *****************************************************************************/
 #include <linux/pm_runtime.h>
+#include <net/tso.h>
 
 #include "iwl-debug.h"
 #include "iwl-csr.h"
@@ -226,6 +227,143 @@ static int iwl_pcie_gen2_set_tb(struct iwl_trans *trans,
 	return idx;
 }
 
+static int iwl_pcie_gen2_build_amsdu(struct iwl_trans *trans,
+				     struct sk_buff *skb,
+				     struct iwl_tfh_tfd *tfd, int start_len,
+				     u8 hdr_len, struct iwl_device_cmd *dev_cmd)
+{
+#ifdef CONFIG_INET
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	struct iwl_tx_cmd *tx_cmd = (void *)dev_cmd->payload;
+	struct ieee80211_hdr *hdr = (void *)skb->data;
+	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
+	unsigned int mss = skb_shinfo(skb)->gso_size;
+	u16 length, iv_len, amsdu_pad;
+	u8 *start_hdr;
+	struct iwl_tso_hdr_page *hdr_page;
+	struct page **page_ptr;
+	struct tso_t tso;
+
+	/* if the packet is protected, then it must be CCMP or GCMP */
+	iv_len = ieee80211_has_protected(hdr->frame_control) ?
+		IEEE80211_CCMP_HDR_LEN : 0;
+
+	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd),
+			     &dev_cmd->hdr, start_len, 0);
+
+	ip_hdrlen = skb_transport_header(skb) - skb_network_header(skb);
+	snap_ip_tcp_hdrlen = 8 + ip_hdrlen + tcp_hdrlen(skb);
+	total_len = skb->len - snap_ip_tcp_hdrlen - hdr_len - iv_len;
+	amsdu_pad = 0;
+
+	/* total amount of header we may need for this A-MSDU */
+	hdr_room = DIV_ROUND_UP(total_len, mss) *
+		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr)) + iv_len;
+
+	/* Our device supports 9 segments at most, it will fit in 1 page */
+	hdr_page = get_page_hdr(trans, hdr_room);
+	if (!hdr_page)
+		return -ENOMEM;
+
+	get_page(hdr_page->page);
+	start_hdr = hdr_page->pos;
+	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->page_offs);
+	*page_ptr = hdr_page->page;
+	memcpy(hdr_page->pos, skb->data + hdr_len, iv_len);
+	hdr_page->pos += iv_len;
+
+	/*
+	 * Pull the ieee80211 header + IV to be able to use TSO core,
+	 * we will restore it for the tx_status flow.
+	 */
+	skb_pull(skb, hdr_len + iv_len);
+
+	/*
+	 * Remove the length of all the headers that we don't actually
+	 * have in the MPDU by themselves, but that we duplicate into
+	 * all the different MSDUs inside the A-MSDU.
+	 */
+	le16_add_cpu(&tx_cmd->len, -snap_ip_tcp_hdrlen);
+
+	tso_start(skb, &tso);
+
+	while (total_len) {
+		/* this is the data left for this subframe */
+		unsigned int data_left = min_t(unsigned int, mss, total_len);
+		struct sk_buff *csum_skb = NULL;
+		unsigned int tb_len;
+		dma_addr_t tb_phys;
+		struct tcphdr *tcph;
+		u8 *iph, *subf_hdrs_start = hdr_page->pos;
+
+		total_len -= data_left;
+
+		memset(hdr_page->pos, 0, amsdu_pad);
+		hdr_page->pos += amsdu_pad;
+		amsdu_pad = (4 - (sizeof(struct ethhdr) + snap_ip_tcp_hdrlen +
+				  data_left)) & 0x3;
+		ether_addr_copy(hdr_page->pos, ieee80211_get_DA(hdr));
+		hdr_page->pos += ETH_ALEN;
+		ether_addr_copy(hdr_page->pos, ieee80211_get_SA(hdr));
+		hdr_page->pos += ETH_ALEN;
+
+		length = snap_ip_tcp_hdrlen + data_left;
+		*((__be16 *)hdr_page->pos) = cpu_to_be16(length);
+		hdr_page->pos += sizeof(length);
+
+		/*
+		 * This will copy the SNAP as well which will be considered
+		 * as MAC header.
+		 */
+		tso_build_hdr(skb, hdr_page->pos, &tso, data_left, !total_len);
+		iph = hdr_page->pos + 8;
+		tcph = (void *)(iph + ip_hdrlen);
+
+		hdr_page->pos += snap_ip_tcp_hdrlen;
+
+		tb_len = hdr_page->pos - start_hdr;
+		tb_phys = dma_map_single(trans->dev, start_hdr,
+					 tb_len, DMA_TO_DEVICE);
+		if (unlikely(dma_mapping_error(trans->dev, tb_phys))) {
+			dev_kfree_skb(csum_skb);
+			goto out_err;
+		}
+		iwl_pcie_gen2_set_tb(trans, tfd, tb_phys, tb_len);
+		trace_iwlwifi_dev_tx_tso_chunk(trans->dev, start_hdr, tb_len);
+		/* add this subframe's headers' length to the tx_cmd */
+		le16_add_cpu(&tx_cmd->len, hdr_page->pos - subf_hdrs_start);
+
+		/* prepare the start_hdr for the next subframe */
+		start_hdr = hdr_page->pos;
+
+		/* put the payload */
+		while (data_left) {
+			tb_len = min_t(unsigned int, tso.size, data_left);
+			tb_phys = dma_map_single(trans->dev, tso.data,
+						 tb_len, DMA_TO_DEVICE);
+			if (unlikely(dma_mapping_error(trans->dev, tb_phys))) {
+				dev_kfree_skb(csum_skb);
+				goto out_err;
+			}
+			iwl_pcie_gen2_set_tb(trans, tfd, tb_phys, tb_len);
+			trace_iwlwifi_dev_tx_tso_chunk(trans->dev, tso.data,
+						       tb_len);
+
+			data_left -= tb_len;
+			tso_build_data(skb, &tso, tb_len);
+		}
+	}
+
+	/* re -add the WiFi header and IV */
+	skb_push(skb, hdr_len + iv_len);
+
+	return 0;
+
+out_err:
+#endif
+	return -EINVAL;
+}
+
 static
 struct iwl_tfh_tfd *iwl_pcie_gen2_build_tfd(struct iwl_trans *trans,
 					    struct iwl_txq *txq,
@@ -238,15 +376,21 @@ struct iwl_tfh_tfd *iwl_pcie_gen2_build_tfd(struct iwl_trans *trans,
 	struct iwl_tfh_tfd *tfd =
 		iwl_pcie_get_tfd(trans_pcie, txq, txq->write_ptr);
 	dma_addr_t tb_phys;
+	bool amsdu;
 	int i, len, tb1_len, tb2_len, hdr_len;
 	void *tb1_addr;
 
 	memset(tfd, 0, sizeof(*tfd));
 
+	amsdu = ieee80211_is_data_qos(hdr->frame_control) &&
+		(*ieee80211_get_qos_ctl(hdr) &
+		 IEEE80211_QOS_CTL_A_MSDU_PRESENT);
+
 	tb_phys = iwl_pcie_get_first_tb_dma(txq, txq->write_ptr);
 	/* The first TB points to bi-directional DMA data */
-	memcpy(&txq->first_tb_bufs[txq->write_ptr], &dev_cmd->hdr,
-	       IWL_FIRST_TB_SIZE);
+	if (!amsdu)
+		memcpy(&txq->first_tb_bufs[txq->write_ptr], &dev_cmd->hdr,
+		       IWL_FIRST_TB_SIZE);
 
 	iwl_pcie_gen2_set_tb(trans, tfd, tb_phys, IWL_FIRST_TB_SIZE);
 
@@ -262,7 +406,11 @@ struct iwl_tfh_tfd *iwl_pcie_gen2_build_tfd(struct iwl_trans *trans,
 	len = sizeof(struct iwl_tx_cmd_gen2) + sizeof(struct iwl_cmd_header) +
 	      ieee80211_hdrlen(hdr->frame_control) - IWL_FIRST_TB_SIZE;
 
-	tb1_len = ALIGN(len, 4);
+	/* do not align A-MSDU to dword as the subframe header aligns it */
+	if (amsdu)
+		tb1_len = len;
+	else
+		tb1_len = ALIGN(len, 4);
 
 	/* map the data for TB1 */
 	tb1_addr = ((u8 *)&dev_cmd->hdr) + IWL_FIRST_TB_SIZE;
@@ -271,8 +419,24 @@ struct iwl_tfh_tfd *iwl_pcie_gen2_build_tfd(struct iwl_trans *trans,
 		goto out_err;
 	iwl_pcie_gen2_set_tb(trans, tfd, tb_phys, tb1_len);
 
-	/* set up TFD's third entry to point to remainder of skb's head */
 	hdr_len = ieee80211_hdrlen(hdr->frame_control);
+
+	if (amsdu) {
+		if (!iwl_pcie_gen2_build_amsdu(trans, skb, tfd,
+					       tb1_len + IWL_FIRST_TB_SIZE,
+					       hdr_len, dev_cmd))
+			goto out_err;
+
+		/*
+		 * building the A-MSDU might have changed this data, so memcpy
+		 * it now
+		 */
+		memcpy(&txq->first_tb_bufs[txq->write_ptr], &dev_cmd->hdr,
+		       IWL_FIRST_TB_SIZE);
+		return tfd;
+	}
+
+	/* set up TFD's third entry to point to remainder of skb's head */
 	tb2_len = skb_headlen(skb) - hdr_len;
 
 	if (tb2_len > 0) {
@@ -303,10 +467,8 @@ struct iwl_tfh_tfd *iwl_pcie_gen2_build_tfd(struct iwl_trans *trans,
 	}
 
 	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd), &dev_cmd->hdr,
-			     IWL_FIRST_TB_SIZE + tb1_len,
-			     skb->data + hdr_len, tb2_len);
-	trace_iwlwifi_dev_tx_data(trans->dev, skb, hdr_len,
-				  skb->len - hdr_len);
+			     IWL_FIRST_TB_SIZE + tb1_len, hdr_len);
+	trace_iwlwifi_dev_tx_data(trans->dev, skb, hdr_len);
 
 	return tfd;
 
@@ -699,7 +861,7 @@ static int iwl_pcie_gen2_send_hcmd_sync(struct iwl_trans *trans,
 	}
 
 	if (!(cmd->flags & CMD_SEND_IN_RFKILL) &&
-	    test_bit(STATUS_RFKILL, &trans->status)) {
+	    test_bit(STATUS_RFKILL_OPMODE, &trans->status)) {
 		IWL_DEBUG_RF_KILL(trans, "RFKILL in SYNC CMD... no rsp\n");
 		ret = -ERFKILL;
 		goto cancel;
@@ -736,7 +898,7 @@ int iwl_trans_pcie_gen2_send_hcmd(struct iwl_trans *trans,
 				  struct iwl_host_cmd *cmd)
 {
 	if (!(cmd->flags & CMD_SEND_IN_RFKILL) &&
-	    test_bit(STATUS_RFKILL, &trans->status)) {
+	    test_bit(STATUS_RFKILL_OPMODE, &trans->status)) {
 		IWL_DEBUG_RF_KILL(trans, "Dropping CMD 0x%x: RF KILL\n",
 				  cmd->id);
 		return -ERFKILL;
@@ -912,7 +1074,7 @@ int iwl_trans_pcie_dyn_txq_alloc(struct iwl_trans *trans,
 	rsp = (void *)hcmd.resp_pkt->data;
 	qid = le16_to_cpu(rsp->queue_number);
 
-	if (qid > ARRAY_SIZE(trans_pcie->txq)) {
+	if (qid >= ARRAY_SIZE(trans_pcie->txq)) {
 		WARN_ONCE(1, "queue index %d unsupported", qid);
 		ret = -EIO;
 		goto error_free_resp;