flow: virtualize flow cache entry methods

This allows to validate the cached object before returning it.
It also allows to destruct object properly, if the last reference
was held in flow cache. This is also a prepartion for caching
bundles in the flow cache.

In return for virtualizing the methods, we save on:
- not having to regenerate the whole flow cache on policy removal:
  each flow matching a killed policy gets refreshed as the getter
  function notices it smartly.
- we do not have to call flow_cache_flush from policy gc, since the
  flow cache now properly deletes the object if it had any references

Signed-off-by: Timo Teras <timo.teras@iki.fi>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

4 files changed, 163 insertions, 102 deletions

diff --git a/include/net/flow.h b/include/net/flow.h
index 809970b..bb08692 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -86,11 +86,26 @@ struct flowi {
 
 struct net;
 struct sock;
-typedef int (*flow_resolve_t)(struct net *net, struct flowi *key, u16 family,
-			      u8 dir, void **objp, atomic_t **obj_refp);
+struct flow_cache_ops;
+
+struct flow_cache_object {
+	const struct flow_cache_ops *ops;
+};
+
+struct flow_cache_ops {
+	struct flow_cache_object *(*get)(struct flow_cache_object *);
+	int (*check)(struct flow_cache_object *);
+	void (*delete)(struct flow_cache_object *);
+};
+
+typedef struct flow_cache_object *(*flow_resolve_t)(
+		struct net *net, struct flowi *key, u16 family,
+		u8 dir, struct flow_cache_object *oldobj, void *ctx);
+
+extern struct flow_cache_object *flow_cache_lookup(
+		struct net *net, struct flowi *key, u16 family,
+		u8 dir, flow_resolve_t resolver, void *ctx);
 
-extern void *flow_cache_lookup(struct net *net, struct flowi *key, u16 family,
-			       u8 dir, flow_resolve_t resolver);
 extern void flow_cache_flush(void);
 extern atomic_t flow_cache_genid;
 
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index d74e080b..35396e2 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -19,6 +19,7 @@
 #include <net/route.h>
 #include <net/ipv6.h>
 #include <net/ip6_fib.h>
+#include <net/flow.h>
 
 #include <linux/interrupt.h>
 
@@ -481,6 +482,7 @@ struct xfrm_policy {
 	atomic_t		refcnt;
 	struct timer_list	timer;
 
+	struct flow_cache_object flo;
 	u32			priority;
 	u32			index;
 	struct xfrm_mark	mark;
diff --git a/net/core/flow.c b/net/core/flow.c
index 1d27ca6..521df52 100644
--- a/net/core/flow.c
+++ b/net/core/flow.c
@@ -26,17 +26,16 @@
 #include <linux/security.h>
 
 struct flow_cache_entry {
-	struct flow_cache_entry	*next;
-	u16			family;
-	u8			dir;
-	u32			genid;
-	struct flowi		key;
-	void			*object;
-	atomic_t		*object_ref;
+	struct flow_cache_entry		*next;
+	u16				family;
+	u8				dir;
+	u32				genid;
+	struct flowi			key;
+	struct flow_cache_object	*object;
 };
 
 struct flow_cache_percpu {
-	struct flow_cache_entry **	hash_table;
+	struct flow_cache_entry		**hash_table;
 	int				hash_count;
 	u32				hash_rnd;
 	int				hash_rnd_recalc;
@@ -44,7 +43,7 @@ struct flow_cache_percpu {
 };
 
 struct flow_flush_info {
-	struct flow_cache *		cache;
+	struct flow_cache		*cache;
 	atomic_t			cpuleft;
 	struct completion		completion;
 };
@@ -52,7 +51,7 @@ struct flow_flush_info {
 struct flow_cache {
 	u32				hash_shift;
 	unsigned long			order;
-	struct flow_cache_percpu *	percpu;
+	struct flow_cache_percpu	*percpu;
 	struct notifier_block		hotcpu_notifier;
 	int				low_watermark;
 	int				high_watermark;
@@ -78,12 +77,21 @@ static void flow_cache_new_hashrnd(unsigned long arg)
 	add_timer(&fc->rnd_timer);
 }
 
+static int flow_entry_valid(struct flow_cache_entry *fle)
+{
+	if (atomic_read(&flow_cache_genid) != fle->genid)
+		return 0;
+	if (fle->object && !fle->object->ops->check(fle->object))
+		return 0;
+	return 1;
+}
+
 static void flow_entry_kill(struct flow_cache *fc,
 			    struct flow_cache_percpu *fcp,
 			    struct flow_cache_entry *fle)
 {
 	if (fle->object)
-		atomic_dec(fle->object_ref);
+		fle->object->ops->delete(fle->object);
 	kmem_cache_free(flow_cachep, fle);
 	fcp->hash_count--;
 }
@@ -96,16 +104,18 @@ static void __flow_cache_shrink(struct flow_cache *fc,
 	int i;
 
 	for (i = 0; i < flow_cache_hash_size(fc); i++) {
-		int k = 0;
+		int saved = 0;
 
 		flp = &fcp->hash_table[i];
-		while ((fle = *flp) != NULL && k < shrink_to) {
-			k++;
-			flp = &fle->next;
-		}
 		while ((fle = *flp) != NULL) {
-			*flp = fle->next;
-			flow_entry_kill(fc, fcp, fle);
+			if (saved < shrink_to &&
+			    flow_entry_valid(fle)) {
+				saved++;
+				flp = &fle->next;
+			} else {
+				*flp = fle->next;
+				flow_entry_kill(fc, fcp, fle);
+			}
 		}
 	}
 }
@@ -166,18 +176,21 @@ static int flow_key_compare(struct flowi *key1, struct flowi *key2)
 	return 0;
 }
 
-void *flow_cache_lookup(struct net *net, struct flowi *key, u16 family, u8 dir,
-			flow_resolve_t resolver)
+struct flow_cache_object *
+flow_cache_lookup(struct net *net, struct flowi *key, u16 family, u8 dir,
+		  flow_resolve_t resolver, void *ctx)
 {
 	struct flow_cache *fc = &flow_cache_global;
 	struct flow_cache_percpu *fcp;
 	struct flow_cache_entry *fle, **head;
+	struct flow_cache_object *flo;
 	unsigned int hash;
 
 	local_bh_disable();
 	fcp = per_cpu_ptr(fc->percpu, smp_processor_id());
 
 	fle = NULL;
+	flo = NULL;
 	/* Packet really early in init?  Making flow_cache_init a
 	 * pre-smp initcall would solve this.  --RR */
 	if (!fcp->hash_table)
@@ -185,27 +198,17 @@ void *flow_cache_lookup(struct net *net, struct flowi *key, u16 family, u8 dir,
 
 	if (fcp->hash_rnd_recalc)
 		flow_new_hash_rnd(fc, fcp);
-	hash = flow_hash_code(fc, fcp, key);
 
+	hash = flow_hash_code(fc, fcp, key);
 	head = &fcp->hash_table[hash];
 	for (fle = *head; fle; fle = fle->next) {
 		if (fle->family == family &&
 		    fle->dir == dir &&
-		    flow_key_compare(key, &fle->key) == 0) {
-			if (fle->genid == atomic_read(&flow_cache_genid)) {
-				void *ret = fle->object;
-
-				if (ret)
-					atomic_inc(fle->object_ref);
-				local_bh_enable();
-
-				return ret;
-			}
+		    flow_key_compare(key, &fle->key) == 0)
 			break;
-		}
 	}
 
-	if (!fle) {
+	if (unlikely(!fle)) {
 		if (fcp->hash_count > fc->high_watermark)
 			flow_cache_shrink(fc, fcp);
 
@@ -219,33 +222,39 @@ void *flow_cache_lookup(struct net *net, struct flowi *key, u16 family, u8 dir,
 			fle->object = NULL;
 			fcp->hash_count++;
 		}
+	} else if (likely(fle->genid == atomic_read(&flow_cache_genid))) {
+		flo = fle->object;
+		if (!flo)
+			goto ret_object;
+		flo = flo->ops->get(flo);
+		if (flo)
+			goto ret_object;
+	} else if (fle->object) {
+	        flo = fle->object;
+	        flo->ops->delete(flo);
+	        fle->object = NULL;
 	}
 
 nocache:
-	{
-		int err;
-		void *obj;
-		atomic_t *obj_ref;
-
-		err = resolver(net, key, family, dir, &obj, &obj_ref);
-
-		if (fle && !err) {
-			fle->genid = atomic_read(&flow_cache_genid);
-
-			if (fle->object)
-				atomic_dec(fle->object_ref);
-
-			fle->object = obj;
-			fle->object_ref = obj_ref;
-			if (obj)
-				atomic_inc(fle->object_ref);
-		}
-		local_bh_enable();
-
-		if (err)
-			obj = ERR_PTR(err);
-		return obj;
+	flo = NULL;
+	if (fle) {
+		flo = fle->object;
+		fle->object = NULL;
+	}
+	flo = resolver(net, key, family, dir, flo, ctx);
+	if (fle) {
+		fle->genid = atomic_read(&flow_cache_genid);
+		if (!IS_ERR(flo))
+			fle->object = flo;
+		else
+			fle->genid--;
+	} else {
+		if (flo && !IS_ERR(flo))
+			flo->ops->delete(flo);
 	}
+ret_object:
+	local_bh_enable();
+	return flo;
 }
 
 static void flow_cache_flush_tasklet(unsigned long data)
@@ -261,13 +270,12 @@ static void flow_cache_flush_tasklet(unsigned long data)
 
 		fle = fcp->hash_table[i];
 		for (; fle; fle = fle->next) {
-			unsigned genid = atomic_read(&flow_cache_genid);
-
-			if (!fle->object || fle->genid == genid)
+			if (flow_entry_valid(fle))
 				continue;
 
+			if (fle->object)
+				fle->object->ops->delete(fle->object);
 			fle->object = NULL;
-			atomic_dec(fle->object_ref);
 		}
 	}
 
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 82789cf..7722bae 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -216,6 +216,35 @@ expired:
 	xfrm_pol_put(xp);
 }
 
+static struct flow_cache_object *xfrm_policy_flo_get(struct flow_cache_object *flo)
+{
+	struct xfrm_policy *pol = container_of(flo, struct xfrm_policy, flo);
+
+	if (unlikely(pol->walk.dead))
+		flo = NULL;
+	else
+		xfrm_pol_hold(pol);
+
+	return flo;
+}
+
+static int xfrm_policy_flo_check(struct flow_cache_object *flo)
+{
+	struct xfrm_policy *pol = container_of(flo, struct xfrm_policy, flo);
+
+	return !pol->walk.dead;
+}
+
+static void xfrm_policy_flo_delete(struct flow_cache_object *flo)
+{
+	xfrm_pol_put(container_of(flo, struct xfrm_policy, flo));
+}
+
+static const struct flow_cache_ops xfrm_policy_fc_ops = {
+	.get = xfrm_policy_flo_get,
+	.check = xfrm_policy_flo_check,
+	.delete = xfrm_policy_flo_delete,
+};
 
 /* Allocate xfrm_policy. Not used here, it is supposed to be used by pfkeyv2
  * SPD calls.
@@ -236,6 +265,7 @@ struct xfrm_policy *xfrm_policy_alloc(struct net *net, gfp_t gfp)
 		atomic_set(&policy->refcnt, 1);
 		setup_timer(&policy->timer, xfrm_policy_timer,
 				(unsigned long)policy);
+		policy->flo.ops = &xfrm_policy_fc_ops;
 	}
 	return policy;
 }
@@ -269,9 +299,6 @@ static void xfrm_policy_gc_kill(struct xfrm_policy *policy)
 	if (del_timer(&policy->timer))
 		atomic_dec(&policy->refcnt);
 
-	if (atomic_read(&policy->refcnt) > 1)
-		flow_cache_flush();
-
 	xfrm_pol_put(policy);
 }
 
@@ -661,10 +688,8 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u32 mark, u8 type,
 	}
 	write_unlock_bh(&xfrm_policy_lock);
 
-	if (ret && delete) {
-		atomic_inc(&flow_cache_genid);
+	if (ret && delete)
 		xfrm_policy_kill(ret);
-	}
 	return ret;
 }
 EXPORT_SYMBOL(xfrm_policy_bysel_ctx);
@@ -703,10 +728,8 @@ struct xfrm_policy *xfrm_policy_byid(struct net *net, u32 mark, u8 type,
 	}
 	write_unlock_bh(&xfrm_policy_lock);
 
-	if (ret && delete) {
-		atomic_inc(&flow_cache_genid);
+	if (ret && delete)
 		xfrm_policy_kill(ret);
-	}
 	return ret;
 }
 EXPORT_SYMBOL(xfrm_policy_byid);
@@ -822,7 +845,6 @@ int xfrm_policy_flush(struct net *net, u8 type, struct xfrm_audit *audit_info)
 	}
 	if (!cnt)
 		err = -ESRCH;
-	atomic_inc(&flow_cache_genid);
 out:
 	write_unlock_bh(&xfrm_policy_lock);
 	return err;
@@ -976,32 +998,35 @@ fail:
 	return ret;
 }
 
-static int xfrm_policy_lookup(struct net *net, struct flowi *fl, u16 family,
-			      u8 dir, void **objp, atomic_t **obj_refp)
+static struct flow_cache_object *
+xfrm_policy_lookup(struct net *net, struct flowi *fl, u16 family,
+		   u8 dir, struct flow_cache_object *old_obj, void *ctx)
 {
 	struct xfrm_policy *pol;
-	int err = 0;
+
+	if (old_obj)
+		xfrm_pol_put(container_of(old_obj, struct xfrm_policy, flo));
 
 #ifdef CONFIG_XFRM_SUB_POLICY
 	pol = xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_SUB, fl, family, dir);
-	if (IS_ERR(pol)) {
-		err = PTR_ERR(pol);
-		pol = NULL;
-	}
-	if (pol || err)
-		goto end;
+	if (IS_ERR(pol))
+		return ERR_CAST(pol);
+	if (pol)
+		goto found;
 #endif
 	pol = xfrm_policy_lookup_bytype(net, XFRM_POLICY_TYPE_MAIN, fl, family, dir);
-	if (IS_ERR(pol)) {
-		err = PTR_ERR(pol);
-		pol = NULL;
-	}
-#ifdef CONFIG_XFRM_SUB_POLICY
-end:
-#endif
-	if ((*objp = (void *) pol) != NULL)
-		*obj_refp = &pol->refcnt;
-	return err;
+	if (IS_ERR(pol))
+		return ERR_CAST(pol);
+	if (pol)
+		goto found;
+	return NULL;
+
+found:
+	/* Resolver returns two references:
+	 * one for cache and one for caller of flow_cache_lookup() */
+	xfrm_pol_hold(pol);
+
+	return &pol->flo;
 }
 
 static inline int policy_to_flow_dir(int dir)
@@ -1091,8 +1116,6 @@ int xfrm_policy_delete(struct xfrm_policy *pol, int dir)
 	pol = __xfrm_policy_unlink(pol, dir);
 	write_unlock_bh(&xfrm_policy_lock);
 	if (pol) {
-		if (dir < XFRM_POLICY_MAX)
-			atomic_inc(&flow_cache_genid);
 		xfrm_policy_kill(pol);
 		return 0;
 	}
@@ -1578,18 +1601,24 @@ restart:
 	}
 
 	if (!policy) {
+		struct flow_cache_object *flo;
+
 		/* To accelerate a bit...  */
 		if ((dst_orig->flags & DST_NOXFRM) ||
 		    !net->xfrm.policy_count[XFRM_POLICY_OUT])
 			goto nopol;
 
-		policy = flow_cache_lookup(net, fl, dst_orig->ops->family,
-					   dir, xfrm_policy_lookup);
-		err = PTR_ERR(policy);
-		if (IS_ERR(policy)) {
+		flo = flow_cache_lookup(net, fl, dst_orig->ops->family,
+					dir, xfrm_policy_lookup, NULL);
+		err = PTR_ERR(flo);
+		if (IS_ERR(flo)) {
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
 			goto dropdst;
 		}
+		if (flo)
+			policy = container_of(flo, struct xfrm_policy, flo);
+		else
+			policy = NULL;
 	}
 
 	if (!policy)
@@ -1939,9 +1968,16 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		}
 	}
 
-	if (!pol)
-		pol = flow_cache_lookup(net, &fl, family, fl_dir,
-					xfrm_policy_lookup);
+	if (!pol) {
+		struct flow_cache_object *flo;
+
+		flo = flow_cache_lookup(net, &fl, family, fl_dir,
+					xfrm_policy_lookup, NULL);
+		if (IS_ERR_OR_NULL(flo))
+			pol = ERR_CAST(flo);
+		else
+			pol = container_of(flo, struct xfrm_policy, flo);
+	}
 
 	if (IS_ERR(pol)) {
 		XFRM_INC_STATS(net, LINUX_MIB_XFRMINPOLERROR);