1 files changed, 124 insertions, 55 deletions

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index eb7d33b41e71..2112146092bf 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -58,6 +58,7 @@
 #include <linux/scatterlist.h>
 #include <linux/errqueue.h>
 #include <linux/prefetch.h>
+#include <linux/bitfield.h>
 #include <linux/if_vlan.h>
 #include <linux/mpls.h>
 #include <linux/kcov.h>
@@ -72,6 +73,7 @@
 #include <net/mptcp.h>
 #include <net/mctp.h>
 #include <net/page_pool.h>
+#include <net/dropreason.h>
 
 #include <linux/uaccess.h>
 #include <trace/events/skb.h>
@@ -122,11 +124,59 @@ EXPORT_SYMBOL(sysctl_max_skb_frags);
 
 #undef FN
 #define FN(reason) [SKB_DROP_REASON_##reason] = #reason,
-const char * const drop_reasons[] = {
+static const char * const drop_reasons[] = {
 	[SKB_CONSUMED] = "CONSUMED",
 	DEFINE_DROP_REASON(FN, FN)
 };
-EXPORT_SYMBOL(drop_reasons);
+
+static const struct drop_reason_list drop_reasons_core = {
+	.reasons = drop_reasons,
+	.n_reasons = ARRAY_SIZE(drop_reasons),
+};
+
+const struct drop_reason_list __rcu *
+drop_reasons_by_subsys[SKB_DROP_REASON_SUBSYS_NUM] = {
+	[SKB_DROP_REASON_SUBSYS_CORE] = RCU_INITIALIZER(&drop_reasons_core),
+};
+EXPORT_SYMBOL(drop_reasons_by_subsys);
+
+/**
+ * drop_reasons_register_subsys - register another drop reason subsystem
+ * @subsys: the subsystem to register, must not be the core
+ * @list: the list of drop reasons within the subsystem, must point to
+ *	a statically initialized list
+ */
+void drop_reasons_register_subsys(enum skb_drop_reason_subsys subsys,
+				  const struct drop_reason_list *list)
+{
+	if (WARN(subsys <= SKB_DROP_REASON_SUBSYS_CORE ||
+		 subsys >= ARRAY_SIZE(drop_reasons_by_subsys),
+		 "invalid subsystem %d\n", subsys))
+		return;
+
+	/* must point to statically allocated memory, so INIT is OK */
+	RCU_INIT_POINTER(drop_reasons_by_subsys[subsys], list);
+}
+EXPORT_SYMBOL_GPL(drop_reasons_register_subsys);
+
+/**
+ * drop_reasons_unregister_subsys - unregister a drop reason subsystem
+ * @subsys: the subsystem to remove, must not be the core
+ *
+ * Note: This will synchronize_rcu() to ensure no users when it returns.
+ */
+void drop_reasons_unregister_subsys(enum skb_drop_reason_subsys subsys)
+{
+	if (WARN(subsys <= SKB_DROP_REASON_SUBSYS_CORE ||
+		 subsys >= ARRAY_SIZE(drop_reasons_by_subsys),
+		 "invalid subsystem %d\n", subsys))
+		return;
+
+	RCU_INIT_POINTER(drop_reasons_by_subsys[subsys], NULL);
+
+	synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(drop_reasons_unregister_subsys);
 
 /**
  *	skb_panic - private function for out-of-line support
@@ -420,10 +470,9 @@ struct sk_buff *build_skb(void *data, unsigned int frag_size)
 {
 	struct sk_buff *skb = __build_skb(data, frag_size);
 
-	if (skb && frag_size) {
+	if (likely(skb && frag_size)) {
 		skb->head_frag = 1;
-		if (page_is_pfmemalloc(virt_to_head_page(data)))
-			skb->pfmemalloc = 1;
+		skb_propagate_pfmemalloc(virt_to_head_page(data), skb);
 	}
 	return skb;
 }
@@ -445,8 +494,7 @@ struct sk_buff *build_skb_around(struct sk_buff *skb,
 
 	if (frag_size) {
 		skb->head_frag = 1;
-		if (page_is_pfmemalloc(virt_to_head_page(data)))
-			skb->pfmemalloc = 1;
+		skb_propagate_pfmemalloc(virt_to_head_page(data), skb);
 	}
 	return skb;
 }
@@ -517,18 +565,16 @@ static void *kmalloc_reserve(unsigned int *size, gfp_t flags, int node,
 #ifdef HAVE_SKB_SMALL_HEAD_CACHE
 	if (obj_size <= SKB_SMALL_HEAD_CACHE_SIZE &&
 	    !(flags & KMALLOC_NOT_NORMAL_BITS)) {
-
-		/* skb_small_head_cache has non power of two size,
-		 * likely forcing SLUB to use order-3 pages.
-		 * We deliberately attempt a NOMEMALLOC allocation only.
-		 */
 		obj = kmem_cache_alloc_node(skb_small_head_cache,
 				flags | __GFP_NOMEMALLOC | __GFP_NOWARN,
 				node);
-		if (obj) {
-			*size = SKB_SMALL_HEAD_CACHE_SIZE;
+		*size = SKB_SMALL_HEAD_CACHE_SIZE;
+		if (obj || !(gfp_pfmemalloc_allowed(flags)))
 			goto out;
-		}
+		/* Try again but now we are using pfmemalloc reserves */
+		ret_pfmemalloc = true;
+		obj = kmem_cache_alloc_node(skb_small_head_cache, flags, node);
+		goto out;
 	}
 #endif
 	*size = obj_size = kmalloc_size_roundup(obj_size);
@@ -843,11 +889,11 @@ static void skb_clone_fraglist(struct sk_buff *skb)
 		skb_get(list);
 }
 
-static bool skb_pp_recycle(struct sk_buff *skb, void *data)
+static bool skb_pp_recycle(struct sk_buff *skb, void *data, bool napi_safe)
 {
 	if (!IS_ENABLED(CONFIG_PAGE_POOL) || !skb->pp_recycle)
 		return false;
-	return page_pool_return_skb_page(virt_to_page(data));
+	return page_pool_return_skb_page(virt_to_page(data), napi_safe);
 }
 
 static void skb_kfree_head(void *head, unsigned int end_offset)
@@ -860,12 +906,12 @@ static void skb_kfree_head(void *head, unsigned int end_offset)
 		kfree(head);
 }
 
-static void skb_free_head(struct sk_buff *skb)
+static void skb_free_head(struct sk_buff *skb, bool napi_safe)
 {
 	unsigned char *head = skb->head;
 
 	if (skb->head_frag) {
-		if (skb_pp_recycle(skb, head))
+		if (skb_pp_recycle(skb, head, napi_safe))
 			return;
 		skb_free_frag(head);
 	} else {
@@ -873,7 +919,8 @@ static void skb_free_head(struct sk_buff *skb)
 	}
 }
 
-static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason)
+static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason,
+			     bool napi_safe)
 {
 	struct skb_shared_info *shinfo = skb_shinfo(skb);
 	int i;
@@ -892,13 +939,13 @@ static void skb_release_data(struct sk_buff *skb, enum skb_drop_reason reason)
 	}
 
 	for (i = 0; i < shinfo->nr_frags; i++)
-		__skb_frag_unref(&shinfo->frags[i], skb->pp_recycle);
+		napi_frag_unref(&shinfo->frags[i], skb->pp_recycle, napi_safe);
 
 free_head:
 	if (shinfo->frag_list)
 		kfree_skb_list_reason(shinfo->frag_list, reason);
 
-	skb_free_head(skb);
+	skb_free_head(skb, napi_safe);
 exit:
 	/* When we clone an SKB we copy the reycling bit. The pp_recycle
 	 * bit is only set on the head though, so in order to avoid races
@@ -959,11 +1006,12 @@ void skb_release_head_state(struct sk_buff *skb)
 }
 
 /* Free everything but the sk_buff shell. */
-static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason)
+static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason,
+			    bool napi_safe)
 {
 	skb_release_head_state(skb);
 	if (likely(skb->head))
-		skb_release_data(skb, reason);
+		skb_release_data(skb, reason, napi_safe);
 }
 
 /**
@@ -977,7 +1025,7 @@ static void skb_release_all(struct sk_buff *skb, enum skb_drop_reason reason)
 
 void __kfree_skb(struct sk_buff *skb)
 {
-	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED);
+	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED, false);
 	kfree_skbmem(skb);
 }
 EXPORT_SYMBOL(__kfree_skb);
@@ -988,7 +1036,10 @@ bool __kfree_skb_reason(struct sk_buff *skb, enum skb_drop_reason reason)
 	if (unlikely(!skb_unref(skb)))
 		return false;
 
-	DEBUG_NET_WARN_ON_ONCE(reason <= 0 || reason >= SKB_DROP_REASON_MAX);
+	DEBUG_NET_WARN_ON_ONCE(reason == SKB_NOT_DROPPED_YET ||
+			       u32_get_bits(reason,
+					    SKB_DROP_REASON_SUBSYS_MASK) >=
+				SKB_DROP_REASON_SUBSYS_NUM);
 
 	if (reason == SKB_CONSUMED)
 		trace_consume_skb(skb, __builtin_return_address(0));
@@ -1031,7 +1082,7 @@ static void kfree_skb_add_bulk(struct sk_buff *skb,
 		return;
 	}
 
-	skb_release_all(skb, reason);
+	skb_release_all(skb, reason, false);
 	sa->skb_array[sa->skb_count++] = skb;
 
 	if (unlikely(sa->skb_count == KFREE_SKB_BULK_SIZE)) {
@@ -1205,7 +1256,7 @@ EXPORT_SYMBOL(consume_skb);
 void __consume_stateless_skb(struct sk_buff *skb)
 {
 	trace_consume_skb(skb, __builtin_return_address(0));
-	skb_release_data(skb, SKB_CONSUMED);
+	skb_release_data(skb, SKB_CONSUMED, false);
 	kfree_skbmem(skb);
 }
 
@@ -1228,9 +1279,9 @@ static void napi_skb_cache_put(struct sk_buff *skb)
 	}
 }
 
-void __kfree_skb_defer(struct sk_buff *skb)
+void __napi_kfree_skb(struct sk_buff *skb, enum skb_drop_reason reason)
 {
-	skb_release_all(skb, SKB_DROP_REASON_NOT_SPECIFIED);
+	skb_release_all(skb, reason, true);
 	napi_skb_cache_put(skb);
 }
 
@@ -1268,7 +1319,7 @@ void napi_consume_skb(struct sk_buff *skb, int budget)
 		return;
 	}
 
-	skb_release_all(skb, SKB_CONSUMED);
+	skb_release_all(skb, SKB_CONSUMED, !!budget);
 	napi_skb_cache_put(skb);
 }
 EXPORT_SYMBOL(napi_consume_skb);
@@ -1399,7 +1450,7 @@ EXPORT_SYMBOL_GPL(alloc_skb_for_msg);
  */
 struct sk_buff *skb_morph(struct sk_buff *dst, struct sk_buff *src)
 {
-	skb_release_all(dst, SKB_CONSUMED);
+	skb_release_all(dst, SKB_CONSUMED, false);
 	return __skb_clone(dst, src);
 }
 EXPORT_SYMBOL_GPL(skb_morph);
@@ -2022,9 +2073,9 @@ int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
 		if (skb_has_frag_list(skb))
 			skb_clone_fraglist(skb);
 
-		skb_release_data(skb, SKB_CONSUMED);
+		skb_release_data(skb, SKB_CONSUMED, false);
 	} else {
-		skb_free_head(skb);
+		skb_free_head(skb, false);
 	}
 	off = (data + nhead) - skb->head;
 
@@ -2082,6 +2133,7 @@ struct sk_buff *skb_realloc_headroom(struct sk_buff *skb, unsigned int headroom)
 }
 EXPORT_SYMBOL(skb_realloc_headroom);
 
+/* Note: We plan to rework this in linux-6.4 */
 int __skb_unclone_keeptruesize(struct sk_buff *skb, gfp_t pri)
 {
 	unsigned int saved_end_offset, saved_truesize;
@@ -2100,6 +2152,22 @@ int __skb_unclone_keeptruesize(struct sk_buff *skb, gfp_t pri)
 	if (likely(skb_end_offset(skb) == saved_end_offset))
 		return 0;
 
+#ifdef HAVE_SKB_SMALL_HEAD_CACHE
+	/* We can not change skb->end if the original or new value
+	 * is SKB_SMALL_HEAD_HEADROOM, as it might break skb_kfree_head().
+	 */
+	if (saved_end_offset == SKB_SMALL_HEAD_HEADROOM ||
+	    skb_end_offset(skb) == SKB_SMALL_HEAD_HEADROOM) {
+		/* We think this path should not be taken.
+		 * Add a temporary trace to warn us just in case.
+		 */
+		pr_err_once("__skb_unclone_keeptruesize() skb_end_offset() %u -> %u\n",
+			    saved_end_offset, skb_end_offset(skb));
+		WARN_ON_ONCE(1);
+		return 0;
+	}
+#endif
+
 	shinfo = skb_shinfo(skb);
 
 	/* We are about to change back skb->end,
@@ -5147,6 +5215,9 @@ void __skb_tstamp_tx(struct sk_buff *orig_skb,
 			skb = alloc_skb(0, GFP_ATOMIC);
 	} else {
 		skb = skb_clone(orig_skb, GFP_ATOMIC);
+
+		if (skb_orphan_frags_rx(skb, GFP_ATOMIC))
+			return;
 	}
 	if (!skb)
 		return;
@@ -5174,6 +5245,7 @@ void skb_tstamp_tx(struct sk_buff *orig_skb,
 }
 EXPORT_SYMBOL_GPL(skb_tstamp_tx);
 
+#ifdef CONFIG_WIRELESS
 void skb_complete_wifi_ack(struct sk_buff *skb, bool acked)
 {
 	struct sock *sk = skb->sk;
@@ -5199,6 +5271,7 @@ void skb_complete_wifi_ack(struct sk_buff *skb, bool acked)
 		kfree_skb(skb);
 }
 EXPORT_SYMBOL_GPL(skb_complete_wifi_ack);
+#endif /* CONFIG_WIRELESS */
 
 /**
  * skb_partial_csum_set - set up and verify partial csum values for packet
@@ -5584,18 +5657,18 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from,
 	if (skb_cloned(to))
 		return false;
 
-	/* In general, avoid mixing slab allocated and page_pool allocated
-	 * pages within the same SKB. However when @to is not pp_recycle and
-	 * @from is cloned, we can transition frag pages from page_pool to
-	 * reference counted.
-	 *
-	 * On the other hand, don't allow coalescing two pp_recycle SKBs if
-	 * @from is cloned, in case the SKB is using page_pool fragment
+	/* In general, avoid mixing page_pool and non-page_pool allocated
+	 * pages within the same SKB. Additionally avoid dealing with clones
+	 * with page_pool pages, in case the SKB is using page_pool fragment
 	 * references (PP_FLAG_PAGE_FRAG). Since we only take full page
 	 * references for cloned SKBs at the moment that would result in
 	 * inconsistent reference counts.
+	 * In theory we could take full references if @from is cloned and
+	 * !@to->pp_recycle but its tricky (due to potential race with
+	 * the clone disappearing) and rare, so not worth dealing with.
 	 */
-	if (to->pp_recycle != (from->pp_recycle && !skb_cloned(from)))
+	if (to->pp_recycle != from->pp_recycle ||
+	    (from->pp_recycle && skb_cloned(from)))
 		return false;
 
 	if (len <= skb_tailroom(to)) {
@@ -5926,7 +5999,6 @@ EXPORT_SYMBOL(skb_ensure_writable);
  */
 int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci)
 {
-	struct vlan_hdr *vhdr;
 	int offset = skb->data - skb_mac_header(skb);
 	int err;
 
@@ -5942,13 +6014,8 @@ int __skb_vlan_pop(struct sk_buff *skb, u16 *vlan_tci)
 
 	skb_postpull_rcsum(skb, skb->data + (2 * ETH_ALEN), VLAN_HLEN);
 
-	vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN);
-	*vlan_tci = ntohs(vhdr->h_vlan_TCI);
-
-	memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN);
-	__skb_pull(skb, VLAN_HLEN);
+	vlan_remove_tag(skb, vlan_tci);
 
-	vlan_set_encap_proto(skb, vhdr);
 	skb->mac_header += VLAN_HLEN;
 
 	if (skb_network_offset(skb) < ETH_HLEN)
@@ -6376,12 +6443,12 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off,
 			skb_frag_ref(skb, i);
 		if (skb_has_frag_list(skb))
 			skb_clone_fraglist(skb);
-		skb_release_data(skb, SKB_CONSUMED);
+		skb_release_data(skb, SKB_CONSUMED, false);
 	} else {
 		/* we can reuse existing recount- all we did was
 		 * relocate values
 		 */
-		skb_free_head(skb);
+		skb_free_head(skb, false);
 	}
 
 	skb->head = data;
@@ -6516,7 +6583,7 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off,
 		skb_kfree_head(data, size);
 		return -ENOMEM;
 	}
-	skb_release_data(skb, SKB_CONSUMED);
+	skb_release_data(skb, SKB_CONSUMED, false);
 
 	skb->head = data;
 	skb->head_frag = 0;
@@ -6800,7 +6867,6 @@ void skb_attempt_defer_free(struct sk_buff *skb)
 {
 	int cpu = skb->alloc_cpu;
 	struct softnet_data *sd;
-	unsigned long flags;
 	unsigned int defer_max;
 	bool kick;
 
@@ -6811,12 +6877,15 @@ nodefer:	__kfree_skb(skb);
 		return;
 	}
 
+	DEBUG_NET_WARN_ON_ONCE(skb_dst(skb));
+	DEBUG_NET_WARN_ON_ONCE(skb->destructor);
+
 	sd = &per_cpu(softnet_data, cpu);
 	defer_max = READ_ONCE(sysctl_skb_defer_max);
 	if (READ_ONCE(sd->defer_count) >= defer_max)
 		goto nodefer;
 
-	spin_lock_irqsave(&sd->defer_lock, flags);
+	spin_lock_bh(&sd->defer_lock);
 	/* Send an IPI every time queue reaches half capacity. */
 	kick = sd->defer_count == (defer_max >> 1);
 	/* Paired with the READ_ONCE() few lines above */
@@ -6825,7 +6894,7 @@ nodefer:	__kfree_skb(skb);
 	skb->next = sd->defer_list;
 	/* Paired with READ_ONCE() in skb_defer_free_flush() */
 	WRITE_ONCE(sd->defer_list, skb);
-	spin_unlock_irqrestore(&sd->defer_lock, flags);
+	spin_unlock_bh(&sd->defer_lock);
 
 	/* Make sure to trigger NET_RX_SOFTIRQ on the remote CPU
 	 * if we are unlucky enough (this seems very unlikely).