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

Pull networking updates from David Miller:

 1) Support IPV6 RA Captive Portal Identifier, from Maciej Żenczykowski.

 2) Use bio_vec in the networking instead of custom skb_frag_t, from
    Matthew Wilcox.

 3) Make use of xmit_more in r8169 driver, from Heiner Kallweit.

 4) Add devmap_hash to xdp, from Toke Høiland-Jørgensen.

 5) Support all variants of 5750X bnxt_en chips, from Michael Chan.

 6) More RTNL avoidance work in the core and mlx5 driver, from Vlad
    Buslov.

 7) Add TCP syn cookies bpf helper, from Petar Penkov.

 8) Add 'nettest' to selftests and use it, from David Ahern.

 9) Add extack support to drop_monitor, add packet alert mode and
    support for HW drops, from Ido Schimmel.

10) Add VLAN offload to stmmac, from Jose Abreu.

11) Lots of devm_platform_ioremap_resource() conversions, from
    YueHaibing.

12) Add IONIC driver, from Shannon Nelson.

13) Several kTLS cleanups, from Jakub Kicinski.

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1930 commits)
  mlxsw: spectrum_buffers: Add the ability to query the CPU port's shared buffer
  mlxsw: spectrum: Register CPU port with devlink
  mlxsw: spectrum_buffers: Prevent changing CPU port's configuration
  net: ena: fix incorrect update of intr_delay_resolution
  net: ena: fix retrieval of nonadaptive interrupt moderation intervals
  net: ena: fix update of interrupt moderation register
  net: ena: remove all old adaptive rx interrupt moderation code from ena_com
  net: ena: remove ena_restore_ethtool_params() and relevant fields
  net: ena: remove old adaptive interrupt moderation code from ena_netdev
  net: ena: remove code duplication in ena_com_update_nonadaptive_moderation_interval _*()
  net: ena: enable the interrupt_moderation in driver_supported_features
  net: ena: reimplement set/get_coalesce()
  net: ena: switch to dim algorithm for rx adaptive interrupt moderation
  net: ena: add intr_moder_rx_interval to struct ena_com_dev and use it
  net: phy: adin: implement Energy Detect Powerdown mode via phy-tunable
  ethtool: implement Energy Detect Powerdown support via phy-tunable
  xen-netfront: do not assume sk_buff_head list is empty in error handling
  s390/ctcm: Delete unnecessary checks before the macro call “dev_kfree_skb”
  net: ena: don't wake up tx queue when down
  drop_monitor: Better sanitize notified packets
  ...

1 files changed, 65 insertions, 12 deletions

diff --git a/net/core/dev.c b/net/core/dev.c
index 5156c0edebe8..71b18e80389f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3467,18 +3467,22 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
 	qdisc_calculate_pkt_len(skb, q);
 
 	if (q->flags & TCQ_F_NOLOCK) {
-		if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED, &q->state))) {
-			__qdisc_drop(skb, &to_free);
-			rc = NET_XMIT_DROP;
-		} else if ((q->flags & TCQ_F_CAN_BYPASS) && q->empty &&
-			   qdisc_run_begin(q)) {
+		if ((q->flags & TCQ_F_CAN_BYPASS) && q->empty &&
+		    qdisc_run_begin(q)) {
+			if (unlikely(test_bit(__QDISC_STATE_DEACTIVATED,
+					      &q->state))) {
+				__qdisc_drop(skb, &to_free);
+				rc = NET_XMIT_DROP;
+				goto end_run;
+			}
 			qdisc_bstats_cpu_update(q, skb);
 
+			rc = NET_XMIT_SUCCESS;
 			if (sch_direct_xmit(skb, q, dev, txq, NULL, true))
 				__qdisc_run(q);
 
+end_run:
 			qdisc_run_end(q);
-			rc = NET_XMIT_SUCCESS;
 		} else {
 			rc = q->enqueue(skb, q, &to_free) & NET_XMIT_MASK;
 			qdisc_run(q);
@@ -3963,6 +3967,8 @@ int dev_weight_rx_bias __read_mostly = 1;  /* bias for backlog weight */
 int dev_weight_tx_bias __read_mostly = 1;  /* bias for output_queue quota */
 int dev_rx_weight __read_mostly = 64;
 int dev_tx_weight __read_mostly = 64;
+/* Maximum number of GRO_NORMAL skbs to batch up for list-RX */
+int gro_normal_batch __read_mostly = 8;
 
 /* Called with irq disabled */
 static inline void ____napi_schedule(struct softnet_data *sd,
@@ -5486,7 +5492,7 @@ static void gro_pull_from_frag0(struct sk_buff *skb, int grow)
 	skb->data_len -= grow;
 	skb->tail += grow;
 
-	pinfo->frags[0].page_offset += grow;
+	skb_frag_off_add(&pinfo->frags[0], grow);
 	skb_frag_size_sub(&pinfo->frags[0], grow);
 
 	if (unlikely(!skb_frag_size(&pinfo->frags[0]))) {
@@ -5747,6 +5753,26 @@ struct sk_buff *napi_get_frags(struct napi_struct *napi)
 }
 EXPORT_SYMBOL(napi_get_frags);
 
+/* Pass the currently batched GRO_NORMAL SKBs up to the stack. */
+static void gro_normal_list(struct napi_struct *napi)
+{
+	if (!napi->rx_count)
+		return;
+	netif_receive_skb_list_internal(&napi->rx_list);
+	INIT_LIST_HEAD(&napi->rx_list);
+	napi->rx_count = 0;
+}
+
+/* Queue one GRO_NORMAL SKB up for list processing.  If batch size exceeded,
+ * pass the whole batch up to the stack.
+ */
+static void gro_normal_one(struct napi_struct *napi, struct sk_buff *skb)
+{
+	list_add_tail(&skb->list, &napi->rx_list);
+	if (++napi->rx_count >= gro_normal_batch)
+		gro_normal_list(napi);
+}
+
 static gro_result_t napi_frags_finish(struct napi_struct *napi,
 				      struct sk_buff *skb,
 				      gro_result_t ret)
@@ -5756,8 +5782,8 @@ static gro_result_t napi_frags_finish(struct napi_struct *napi,
 	case GRO_HELD:
 		__skb_push(skb, ETH_HLEN);
 		skb->protocol = eth_type_trans(skb, skb->dev);
-		if (ret == GRO_NORMAL && netif_receive_skb_internal(skb))
-			ret = GRO_DROP;
+		if (ret == GRO_NORMAL)
+			gro_normal_one(napi, skb);
 		break;
 
 	case GRO_DROP:
@@ -6034,6 +6060,8 @@ bool napi_complete_done(struct napi_struct *n, int work_done)
 				 NAPIF_STATE_IN_BUSY_POLL)))
 		return false;
 
+	gro_normal_list(n);
+
 	if (n->gro_bitmask) {
 		unsigned long timeout = 0;
 
@@ -6119,10 +6147,19 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock)
 	 * Ideally, a new ndo_busy_poll_stop() could avoid another round.
 	 */
 	rc = napi->poll(napi, BUSY_POLL_BUDGET);
+	/* We can't gro_normal_list() here, because napi->poll() might have
+	 * rearmed the napi (napi_complete_done()) in which case it could
+	 * already be running on another CPU.
+	 */
 	trace_napi_poll(napi, rc, BUSY_POLL_BUDGET);
 	netpoll_poll_unlock(have_poll_lock);
-	if (rc == BUSY_POLL_BUDGET)
+	if (rc == BUSY_POLL_BUDGET) {
+		/* As the whole budget was spent, we still own the napi so can
+		 * safely handle the rx_list.
+		 */
+		gro_normal_list(napi);
 		__napi_schedule(napi);
+	}
 	local_bh_enable();
 }
 
@@ -6167,6 +6204,7 @@ restart:
 		}
 		work = napi_poll(napi, BUSY_POLL_BUDGET);
 		trace_napi_poll(napi, work, BUSY_POLL_BUDGET);
+		gro_normal_list(napi);
 count:
 		if (work > 0)
 			__NET_ADD_STATS(dev_net(napi->dev),
@@ -6272,6 +6310,8 @@ void netif_napi_add(struct net_device *dev, struct napi_struct *napi,
 	napi->timer.function = napi_watchdog;
 	init_gro_hash(napi);
 	napi->skb = NULL;
+	INIT_LIST_HEAD(&napi->rx_list);
+	napi->rx_count = 0;
 	napi->poll = poll;
 	if (weight > NAPI_POLL_WEIGHT)
 		netdev_err_once(dev, "%s() called with weight %d\n", __func__,
@@ -6368,6 +6408,8 @@ static int napi_poll(struct napi_struct *n, struct list_head *repoll)
 		goto out_unlock;
 	}
 
+	gro_normal_list(n);
+
 	if (n->gro_bitmask) {
 		/* flush too old packets
 		 * If HZ < 1000, flush all packets.
@@ -8088,12 +8130,15 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
 		bpf_chk = generic_xdp_install;
 
 	if (fd >= 0) {
+		u32 prog_id;
+
 		if (!offload && __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG)) {
 			NL_SET_ERR_MSG(extack, "native and generic XDP can't be active at the same time");
 			return -EEXIST;
 		}
-		if ((flags & XDP_FLAGS_UPDATE_IF_NOEXIST) &&
-		    __dev_xdp_query(dev, bpf_op, query)) {
+
+		prog_id = __dev_xdp_query(dev, bpf_op, query);
+		if ((flags & XDP_FLAGS_UPDATE_IF_NOEXIST) && prog_id) {
 			NL_SET_ERR_MSG(extack, "XDP program already attached");
 			return -EBUSY;
 		}
@@ -8108,6 +8153,14 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
 			bpf_prog_put(prog);
 			return -EINVAL;
 		}
+
+		if (prog->aux->id == prog_id) {
+			bpf_prog_put(prog);
+			return 0;
+		}
+	} else {
+		if (!__dev_xdp_query(dev, bpf_op, query))
+			return 0;
 	}
 
 	err = dev_xdp_install(dev, bpf_op, extack, flags, prog);