These changes are the raw update to linux-4.4.6-rt14. Kernel sources

[kvmfornfv.git] / kernel / net / core / dst.c

diff --git a/kernel/net/core/dst.c b/kernel/net/core/dst.c
index e956ce6..a1656e3 100644 (file)
--- a/kernel/net/core/dst.c
+++ b/kernel/net/core/dst.c
@@ -20,8 +20,10 @@
 #include <net/net_namespace.h>
 #include <linux/sched.h>
 #include <linux/prefetch.h>
+#include <net/lwtunnel.h>
 
 #include <net/dst.h>
+#include <net/dst_metadata.h>
 
 /*
  * Theory of operations:
@@ -142,12 +144,12 @@ loop:
        mutex_unlock(&dst_gc_mutex);
 }
 
-int dst_discard_sk(struct sock *sk, struct sk_buff *skb)
+int dst_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb)
 {
        kfree_skb(skb);
        return 0;
 }
-EXPORT_SYMBOL(dst_discard_sk);
+EXPORT_SYMBOL(dst_discard_out);
 
 const u32 dst_default_metrics[RTAX_MAX + 1] = {
        /* This initializer is needed to force linker to place this variable
@@ -158,19 +160,10 @@ const u32 dst_default_metrics[RTAX_MAX + 1] = {
        [RTAX_MAX] = 0xdeadbeef,
 };
 
-
-void *dst_alloc(struct dst_ops *ops, struct net_device *dev,
-               int initial_ref, int initial_obsolete, unsigned short flags)
+void dst_init(struct dst_entry *dst, struct dst_ops *ops,
+             struct net_device *dev, int initial_ref, int initial_obsolete,
+             unsigned short flags)
 {
-       struct dst_entry *dst;
-
-       if (ops->gc && dst_entries_get_fast(ops) > ops->gc_thresh) {
-               if (ops->gc(ops))
-                       return NULL;
-       }
-       dst = kmem_cache_alloc(ops->kmem_cachep, GFP_ATOMIC);
-       if (!dst)
-               return NULL;
        dst->child = NULL;
        dst->dev = dev;
        if (dev)
@@ -184,7 +177,7 @@ void *dst_alloc(struct dst_ops *ops, struct net_device *dev,
        dst->xfrm = NULL;
 #endif
        dst->input = dst_discard;
-       dst->output = dst_discard_sk;
+       dst->output = dst_discard_out;
        dst->error = 0;
        dst->obsolete = initial_obsolete;
        dst->header_len = 0;
@@ -192,6 +185,7 @@ void *dst_alloc(struct dst_ops *ops, struct net_device *dev,
 #ifdef CONFIG_IP_ROUTE_CLASSID
        dst->tclassid = 0;
 #endif
+       dst->lwtstate = NULL;
        atomic_set(&dst->__refcnt, initial_ref);
        dst->__use = 0;
        dst->lastuse = jiffies;
@@ -200,6 +194,25 @@ void *dst_alloc(struct dst_ops *ops, struct net_device *dev,
        dst->next = NULL;
        if (!(flags & DST_NOCOUNT))
                dst_entries_add(ops, 1);
+}
+EXPORT_SYMBOL(dst_init);
+
+void *dst_alloc(struct dst_ops *ops, struct net_device *dev,
+               int initial_ref, int initial_obsolete, unsigned short flags)
+{
+       struct dst_entry *dst;
+
+       if (ops->gc && dst_entries_get_fast(ops) > ops->gc_thresh) {
+               if (ops->gc(ops))
+                       return NULL;
+       }
+
+       dst = kmem_cache_alloc(ops->kmem_cachep, GFP_ATOMIC);
+       if (!dst)
+               return NULL;
+
+       dst_init(dst, ops, dev, initial_ref, initial_obsolete, flags);
+
        return dst;
 }
 EXPORT_SYMBOL(dst_alloc);
@@ -211,7 +224,7 @@ static void ___dst_free(struct dst_entry *dst)
         */
        if (dst->dev == NULL || !(dst->dev->flags&IFF_UP)) {
                dst->input = dst_discard;
-               dst->output = dst_discard_sk;
+               dst->output = dst_discard_out;
        }
        dst->obsolete = DST_OBSOLETE_DEAD;
 }
@@ -248,7 +261,13 @@ again:
                dst->ops->destroy(dst);
        if (dst->dev)
                dev_put(dst->dev);
-       kmem_cache_free(dst->ops->kmem_cachep, dst);
+
+       lwtstate_put(dst->lwtstate);
+
+       if (dst->flags & DST_METADATA)
+               kfree(dst);
+       else
+               kmem_cache_free(dst->ops->kmem_cachep, dst);
 
        dst = child;
        if (dst) {
@@ -282,10 +301,13 @@ void dst_release(struct dst_entry *dst)
 {
        if (dst) {
                int newrefcnt;
+               unsigned short nocache = dst->flags & DST_NOCACHE;
 
                newrefcnt = atomic_dec_return(&dst->__refcnt);
-               WARN_ON(newrefcnt < 0);
-               if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt)
+               if (unlikely(newrefcnt < 0))
+                       net_warn_ratelimited("%s: dst:%p refcnt:%d\n",
+                                            __func__, dst, newrefcnt);
+               if (!newrefcnt && unlikely(nocache))
                        call_rcu(&dst->rcu_head, dst_destroy_rcu);
        }
 }
@@ -327,6 +349,69 @@ void __dst_destroy_metrics_generic(struct dst_entry *dst, unsigned long old)
 }
 EXPORT_SYMBOL(__dst_destroy_metrics_generic);
 
+static struct dst_ops md_dst_ops = {
+       .family =               AF_UNSPEC,
+};
+
+static int dst_md_discard_out(struct net *net, struct sock *sk, struct sk_buff *skb)
+{
+       WARN_ONCE(1, "Attempting to call output on metadata dst\n");
+       kfree_skb(skb);
+       return 0;
+}
+
+static int dst_md_discard(struct sk_buff *skb)
+{
+       WARN_ONCE(1, "Attempting to call input on metadata dst\n");
+       kfree_skb(skb);
+       return 0;
+}
+
+static void __metadata_dst_init(struct metadata_dst *md_dst, u8 optslen)
+{
+       struct dst_entry *dst;
+
+       dst = &md_dst->dst;
+       dst_init(dst, &md_dst_ops, NULL, 1, DST_OBSOLETE_NONE,
+                DST_METADATA | DST_NOCACHE | DST_NOCOUNT);
+
+       dst->input = dst_md_discard;
+       dst->output = dst_md_discard_out;
+
+       memset(dst + 1, 0, sizeof(*md_dst) + optslen - sizeof(*dst));
+}
+
+struct metadata_dst *metadata_dst_alloc(u8 optslen, gfp_t flags)
+{
+       struct metadata_dst *md_dst;
+
+       md_dst = kmalloc(sizeof(*md_dst) + optslen, flags);
+       if (!md_dst)
+               return NULL;
+
+       __metadata_dst_init(md_dst, optslen);
+
+       return md_dst;
+}
+EXPORT_SYMBOL_GPL(metadata_dst_alloc);
+
+struct metadata_dst __percpu *metadata_dst_alloc_percpu(u8 optslen, gfp_t flags)
+{
+       int cpu;
+       struct metadata_dst __percpu *md_dst;
+
+       md_dst = __alloc_percpu_gfp(sizeof(struct metadata_dst) + optslen,
+                                   __alignof__(struct metadata_dst), flags);
+       if (!md_dst)
+               return NULL;
+
+       for_each_possible_cpu(cpu)
+               __metadata_dst_init(per_cpu_ptr(md_dst, cpu), optslen);
+
+       return md_dst;
+}
+EXPORT_SYMBOL_GPL(metadata_dst_alloc_percpu);
+
 /* Dirty hack. We did it in 2.2 (in __dst_free),
  * we have _very_ good reasons not to repeat
  * this mistake in 2.3, but we have no choice
@@ -346,7 +431,7 @@ static void dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 
        if (!unregister) {
                dst->input = dst_discard;
-               dst->output = dst_discard_sk;
+               dst->output = dst_discard_out;
        } else {
                dst->dev = dev_net(dst->dev)->loopback_dev;
                dev_hold(dst->dev);
@@ -391,7 +476,7 @@ static struct notifier_block dst_dev_notifier = {
        .priority = -10, /* must be called after other network notifiers */
 };
 
-void __init dst_init(void)
+void __init dst_subsys_init(void)
 {
        register_netdevice_notifier(&dst_dev_notifier);
 }