1 files changed, 32 insertions, 128 deletions

diff --git a/net/core/dev.c b/net/core/dev.c
index 29101c98399f..0921609dfa81 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2403,28 +2403,6 @@ void netif_schedule_queue(struct netdev_queue *txq)
 }
 EXPORT_SYMBOL(netif_schedule_queue);
 
-/**
- *      netif_wake_subqueue - allow sending packets on subqueue
- *      @dev: network device
- *      @queue_index: sub queue index
- *
- * Resume individual transmit queue of a device with multiple transmit queues.
- */
-void netif_wake_subqueue(struct net_device *dev, u16 queue_index)
-{
-        struct netdev_queue *txq = netdev_get_tx_queue(dev, queue_index);
-
-        if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &txq->state)) {
-                struct Qdisc *q;
-
-                rcu_read_lock();
-                q = rcu_dereference(txq->qdisc);
-                __netif_schedule(q);
-                rcu_read_unlock();
-        }
-}
-EXPORT_SYMBOL(netif_wake_subqueue);
-
 void netif_tx_wake_queue(struct netdev_queue *dev_queue)
 {
         if (test_and_clear_bit(__QUEUE_STATE_DRV_XOFF, &dev_queue->state)) {
@@ -2654,9 +2632,10 @@ EXPORT_SYMBOL(skb_mac_gso_segment);
 static inline bool skb_needs_check(struct sk_buff *skb, bool tx_path)
 {
         if (tx_path)
-                return skb->ip_summed != CHECKSUM_PARTIAL;
-        else
-                return skb->ip_summed == CHECKSUM_NONE;
+                return skb->ip_summed != CHECKSUM_PARTIAL &&
+                       skb->ip_summed != CHECKSUM_NONE;
+
+        return skb->ip_summed == CHECKSUM_NONE;
 }
 
 /**
@@ -2675,11 +2654,12 @@ static inline bool skb_needs_check(struct sk_buff *skb, bool tx_path)
 struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
                                   netdev_features_t features, bool tx_path)
 {
+        struct sk_buff *segs;
+
         if (unlikely(skb_needs_check(skb, tx_path))) {
                 int err;
 
-                skb_warn_bad_offload(skb);
-
+                /* We're going to init ->check field in TCP or UDP header */
                 err = skb_cow_head(skb, 0);
                 if (err < 0)
                         return ERR_PTR(err);
@@ -2707,7 +2687,12 @@ struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
         skb_reset_mac_header(skb);
         skb_reset_mac_len(skb);
 
-        return skb_mac_gso_segment(skb, features);
+        segs = skb_mac_gso_segment(skb, features);
+
+        if (unlikely(skb_needs_check(skb, tx_path)))
+                skb_warn_bad_offload(skb);
+
+        return segs;
 }
 EXPORT_SYMBOL(__skb_gso_segment);
 
@@ -3148,9 +3133,7 @@ sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev)
         if (!cl)
                 return skb;
 
-        /* skb->tc_verd and qdisc_skb_cb(skb)->pkt_len were already set
-         * earlier by the caller.
-         */
+        /* qdisc_skb_cb(skb)->pkt_len was already set by the caller. */
         qdisc_bstats_cpu_update(cl->q, skb);
 
         switch (tc_classify(skb, cl, &cl_res, false)) {
@@ -3315,7 +3298,7 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv)
 
         qdisc_pkt_len_init(skb);
 #ifdef CONFIG_NET_CLS_ACT
-        skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_EGRESS);
+        skb->tc_at_ingress = 0;
 # ifdef CONFIG_NET_EGRESS
         if (static_key_false(&egress_needed)) {
                 skb = sch_handle_egress(skb, &rc, dev);
@@ -3422,7 +3405,11 @@ EXPORT_SYMBOL(netdev_max_backlog);
 
 int netdev_tstamp_prequeue __read_mostly = 1;
 int netdev_budget __read_mostly = 300;
-int weight_p __read_mostly = 64;            /* old backlog weight */
+int weight_p __read_mostly = 64;           /* old backlog weight */
+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;
 
 /* Called with irq disabled */
 static inline void ____napi_schedule(struct softnet_data *sd,
@@ -3911,7 +3898,7 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret,
         }
 
         qdisc_skb_cb(skb)->pkt_len = skb->len;
-        skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_INGRESS);
+        skb->tc_at_ingress = 1;
         qdisc_bstats_cpu_update(cl->q, skb);
 
         switch (tc_classify(skb, cl, &cl_res, false)) {
@@ -3976,9 +3963,7 @@ int netdev_rx_handler_register(struct net_device *dev,
                                rx_handler_func_t *rx_handler,
                                void *rx_handler_data)
 {
-        ASSERT_RTNL();
-
-        if (dev->rx_handler)
+        if (netdev_is_rx_handler_busy(dev))
                 return -EBUSY;
 
         /* Note: rx_handler_data must be set before rx_handler */
@@ -4084,12 +4069,8 @@ another_round:
                         goto out;
         }
 
-#ifdef CONFIG_NET_CLS_ACT
-        if (skb->tc_verd & TC_NCLS) {
-                skb->tc_verd = CLR_TC_NCLS(skb->tc_verd);
-                goto ncls;
-        }
-#endif
+        if (skb_skip_tc_classify(skb))
+                goto skip_classify;
 
         if (pfmemalloc)
                 goto skip_taps;
@@ -4117,10 +4098,8 @@ skip_taps:
                         goto out;
         }
 #endif
-#ifdef CONFIG_NET_CLS_ACT
-        skb->tc_verd = 0;
-ncls:
-#endif
+        skb_reset_tc(skb);
+skip_classify:
         if (pfmemalloc && !skb_pfmemalloc_protocol(skb))
                 goto drop;
 
@@ -4616,6 +4595,7 @@ static gro_result_t napi_skb_finish(gro_result_t ret, struct sk_buff *skb)
         case GRO_MERGED_FREE:
                 if (NAPI_GRO_CB(skb)->free == NAPI_GRO_FREE_STOLEN_HEAD) {
                         skb_dst_drop(skb);
+                        secpath_reset(skb);
                         kmem_cache_free(skbuff_head_cache, skb);
                 } else {
                         __kfree_skb(skb);
@@ -4656,6 +4636,7 @@ static void napi_reuse_skb(struct napi_struct *napi, struct sk_buff *skb)
         skb->encapsulation = 0;
         skb_shinfo(skb)->gso_type = 0;
         skb->truesize = SKB_TRUESIZE(skb_end_offset(skb));
+        secpath_reset(skb);
 
         napi->skb = skb;
 }
@@ -4830,7 +4811,7 @@ static int process_backlog(struct napi_struct *napi, int quota)
                 net_rps_action_and_irq_enable(sd);
         }
 
-        napi->weight = weight_p;
+        napi->weight = dev_rx_weight;
         while (again) {
                 struct sk_buff *skb;
 
@@ -4897,23 +4878,6 @@ void __napi_schedule_irqoff(struct napi_struct *n)
 }
 EXPORT_SYMBOL(__napi_schedule_irqoff);
 
-bool __napi_complete(struct napi_struct *n)
-{
-        BUG_ON(!test_bit(NAPI_STATE_SCHED, &n->state));
-
-        /* Some drivers call us directly, instead of calling
-         * napi_complete_done().
-         */
-        if (unlikely(test_bit(NAPI_STATE_IN_BUSY_POLL, &n->state)))
-                return false;
-
-        list_del_init(&n->poll_list);
-        smp_mb__before_atomic();
-        clear_bit(NAPI_STATE_SCHED, &n->state);
-        return true;
-}
-EXPORT_SYMBOL(__napi_complete);
-
 bool napi_complete_done(struct napi_struct *n, int work_done)
 {
         unsigned long flags;
@@ -4940,14 +4904,13 @@ bool napi_complete_done(struct napi_struct *n, int work_done)
                 else
                         napi_gro_flush(n, false);
         }
-        if (likely(list_empty(&n->poll_list))) {
-                WARN_ON_ONCE(!test_and_clear_bit(NAPI_STATE_SCHED, &n->state));
-        } else {
+        if (unlikely(!list_empty(&n->poll_list))) {
                 /* If n->poll_list is not empty, we need to mask irqs */
                 local_irq_save(flags);
-                __napi_complete(n);
+                list_del_init(&n->poll_list);
                 local_irq_restore(flags);
         }
+        WARN_ON_ONCE(!test_and_clear_bit(NAPI_STATE_SCHED, &n->state));
         return true;
 }
 EXPORT_SYMBOL(napi_complete_done);
@@ -4993,7 +4956,6 @@ bool sk_busy_loop(struct sock *sk, int nonblock)
 {
         unsigned long end_time = !nonblock ? sk_busy_loop_end_time(sk) : 0;
         int (*napi_poll)(struct napi_struct *napi, int budget);
-        int (*busy_poll)(struct napi_struct *dev);
         void *have_poll_lock = NULL;
         struct napi_struct *napi;
         int rc;
@@ -5008,17 +4970,10 @@ restart:
         if (!napi)
                 goto out;
 
-        /* Note: ndo_busy_poll method is optional in linux-4.5 */
-        busy_poll = napi->dev->netdev_ops->ndo_busy_poll;
-
         preempt_disable();
         for (;;) {
                 rc = 0;
                 local_bh_disable();
-                if (busy_poll) {
-                        rc = busy_poll(napi);
-                        goto count;
-                }
                 if (!napi_poll) {
                         unsigned long val = READ_ONCE(napi->state);
 
@@ -6151,50 +6106,6 @@ void netdev_lower_state_changed(struct net_device *lower_dev,
 }
 EXPORT_SYMBOL(netdev_lower_state_changed);
 
-int netdev_default_l2upper_neigh_construct(struct net_device *dev,
-                                           struct neighbour *n)
-{
-        struct net_device *lower_dev, *stop_dev;
-        struct list_head *iter;
-        int err;
-
-        netdev_for_each_lower_dev(dev, lower_dev, iter) {
-                if (!lower_dev->netdev_ops->ndo_neigh_construct)
-                        continue;
-                err = lower_dev->netdev_ops->ndo_neigh_construct(lower_dev, n);
-                if (err) {
-                        stop_dev = lower_dev;
-                        goto rollback;
-                }
-        }
-        return 0;
-
-rollback:
-        netdev_for_each_lower_dev(dev, lower_dev, iter) {
-                if (lower_dev == stop_dev)
-                        break;
-                if (!lower_dev->netdev_ops->ndo_neigh_destroy)
-                        continue;
-                lower_dev->netdev_ops->ndo_neigh_destroy(lower_dev, n);
-        }
-        return err;
-}
-EXPORT_SYMBOL_GPL(netdev_default_l2upper_neigh_construct);
-
-void netdev_default_l2upper_neigh_destroy(struct net_device *dev,
-                                          struct neighbour *n)
-{
-        struct net_device *lower_dev;
-        struct list_head *iter;
-
-        netdev_for_each_lower_dev(dev, lower_dev, iter) {
-                if (!lower_dev->netdev_ops->ndo_neigh_destroy)
-                        continue;
-                lower_dev->netdev_ops->ndo_neigh_destroy(lower_dev, n);
-        }
-}
-EXPORT_SYMBOL_GPL(netdev_default_l2upper_neigh_destroy);
-
 static void dev_change_rx_flags(struct net_device *dev, int flags)
 {
         const struct net_device_ops *ops = dev->netdev_ops;
@@ -6971,13 +6882,6 @@ static netdev_features_t netdev_fix_features(struct net_device *dev,
                 features &= ~dev->gso_partial_features;
         }
 
-#ifdef CONFIG_NET_RX_BUSY_POLL
-        if (dev->netdev_ops->ndo_busy_poll)
-                features |= NETIF_F_BUSY_POLL;
-        else
-#endif
-                features &= ~NETIF_F_BUSY_POLL;
-
         return features;
 }