diff options
Diffstat (limited to 'net/sched/act_api.c')
| -rw-r--r-- | net/sched/act_api.c | 723 | 
1 files changed, 342 insertions, 381 deletions
diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 23b25f89e7e..648778aef1a 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -20,53 +20,47 @@  #include <linux/init.h>  #include <linux/kmod.h>  #include <linux/err.h> +#include <linux/module.h>  #include <net/net_namespace.h>  #include <net/sock.h>  #include <net/sch_generic.h>  #include <net/act_api.h>  #include <net/netlink.h> -static void tcf_common_free_rcu(struct rcu_head *head) +void tcf_hash_destroy(struct tc_action *a)  { -	kfree(container_of(head, struct tcf_common, tcfc_rcu)); -} +	struct tcf_common *p = a->priv; +	struct tcf_hashinfo *hinfo = a->ops->hinfo; -void tcf_hash_destroy(struct tcf_common *p, struct tcf_hashinfo *hinfo) -{ -	unsigned int h = tcf_hash(p->tcfc_index, hinfo->hmask); -	struct tcf_common **p1p; - -	for (p1p = &hinfo->htab[h]; *p1p; p1p = &(*p1p)->tcfc_next) { -		if (*p1p == p) { -			write_lock_bh(hinfo->lock); -			*p1p = p->tcfc_next; -			write_unlock_bh(hinfo->lock); -			gen_kill_estimator(&p->tcfc_bstats, -					   &p->tcfc_rate_est); -			/* -			 * gen_estimator est_timer() might access p->tcfc_lock -			 * or bstats, wait a RCU grace period before freeing p -			 */ -			call_rcu(&p->tcfc_rcu, tcf_common_free_rcu); -			return; -		} -	} -	WARN_ON(1); +	spin_lock_bh(&hinfo->lock); +	hlist_del(&p->tcfc_head); +	spin_unlock_bh(&hinfo->lock); +	gen_kill_estimator(&p->tcfc_bstats, +			   &p->tcfc_rate_est); +	/* +	 * gen_estimator est_timer() might access p->tcfc_lock +	 * or bstats, wait a RCU grace period before freeing p +	 */ +	kfree_rcu(p, tcfc_rcu);  }  EXPORT_SYMBOL(tcf_hash_destroy); -int tcf_hash_release(struct tcf_common *p, int bind, -		     struct tcf_hashinfo *hinfo) +int tcf_hash_release(struct tc_action *a, int bind)  { +	struct tcf_common *p = a->priv;  	int ret = 0;  	if (p) {  		if (bind)  			p->tcfc_bindcnt--; +		else if (p->tcfc_bindcnt > 0) +			return -EPERM;  		p->tcfc_refcnt--;  		if (p->tcfc_bindcnt <= 0 && p->tcfc_refcnt <= 0) { -			tcf_hash_destroy(p, hinfo); +			if (a->ops->cleanup) +				a->ops->cleanup(a, bind); +			tcf_hash_destroy(a);  			ret = 1;  		}  	} @@ -75,20 +69,22 @@ int tcf_hash_release(struct tcf_common *p, int bind,  EXPORT_SYMBOL(tcf_hash_release);  static int tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb, -			   struct tc_action *a, struct tcf_hashinfo *hinfo) +			   struct tc_action *a)  { +	struct tcf_hashinfo *hinfo = a->ops->hinfo; +	struct hlist_head *head;  	struct tcf_common *p; -	int err = 0, index = -1,i = 0, s_i = 0, n_i = 0; +	int err = 0, index = -1, i = 0, s_i = 0, n_i = 0;  	struct nlattr *nest; -	read_lock_bh(hinfo->lock); +	spin_lock_bh(&hinfo->lock);  	s_i = cb->args[0];  	for (i = 0; i < (hinfo->hmask + 1); i++) { -		p = hinfo->htab[tcf_hash(i, hinfo->hmask)]; +		head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; -		for (; p; p = p->tcfc_next) { +		hlist_for_each_entry_rcu(p, head, tcfc_head) {  			index++;  			if (index < s_i)  				continue; @@ -111,7 +107,7 @@ static int tcf_dump_walker(struct sk_buff *skb, struct netlink_callback *cb,  		}  	}  done: -	read_unlock_bh(hinfo->lock); +	spin_unlock_bh(&hinfo->lock);  	if (n_i)  		cb->args[0] += n_i;  	return n_i; @@ -121,79 +117,82 @@ nla_put_failure:  	goto done;  } -static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a, -			  struct tcf_hashinfo *hinfo) +static int tcf_del_walker(struct sk_buff *skb, struct tc_action *a)  { -	struct tcf_common *p, *s_p; +	struct tcf_hashinfo *hinfo = a->ops->hinfo; +	struct hlist_head *head; +	struct hlist_node *n; +	struct tcf_common *p;  	struct nlattr *nest; -	int i= 0, n_i = 0; +	int i = 0, n_i = 0; +	int ret = -EINVAL;  	nest = nla_nest_start(skb, a->order);  	if (nest == NULL)  		goto nla_put_failure; -	NLA_PUT_STRING(skb, TCA_KIND, a->ops->kind); +	if (nla_put_string(skb, TCA_KIND, a->ops->kind)) +		goto nla_put_failure;  	for (i = 0; i < (hinfo->hmask + 1); i++) { -		p = hinfo->htab[tcf_hash(i, hinfo->hmask)]; - -		while (p != NULL) { -			s_p = p->tcfc_next; -			if (ACT_P_DELETED == tcf_hash_release(p, 0, hinfo)) -				 module_put(a->ops->owner); -			n_i++; -			p = s_p; +		head = &hinfo->htab[tcf_hash(i, hinfo->hmask)]; +		hlist_for_each_entry_safe(p, n, head, tcfc_head) { +			a->priv = p; +			ret = tcf_hash_release(a, 0); +			if (ret == ACT_P_DELETED) { +				module_put(a->ops->owner); +				n_i++; +			} else if (ret < 0) +				goto nla_put_failure;  		}  	} -	NLA_PUT_U32(skb, TCA_FCNT, n_i); +	if (nla_put_u32(skb, TCA_FCNT, n_i)) +		goto nla_put_failure;  	nla_nest_end(skb, nest);  	return n_i;  nla_put_failure:  	nla_nest_cancel(skb, nest); -	return -EINVAL; +	return ret;  } -int tcf_generic_walker(struct sk_buff *skb, struct netlink_callback *cb, -		       int type, struct tc_action *a) +static int tcf_generic_walker(struct sk_buff *skb, struct netlink_callback *cb, +			      int type, struct tc_action *a)  { -	struct tcf_hashinfo *hinfo = a->ops->hinfo; -  	if (type == RTM_DELACTION) { -		return tcf_del_walker(skb, a, hinfo); +		return tcf_del_walker(skb, a);  	} else if (type == RTM_GETACTION) { -		return tcf_dump_walker(skb, cb, a, hinfo); +		return tcf_dump_walker(skb, cb, a);  	} else {  		WARN(1, "tcf_generic_walker: unknown action %d\n", type);  		return -EINVAL;  	}  } -EXPORT_SYMBOL(tcf_generic_walker); -struct tcf_common *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo) +static struct tcf_common *tcf_hash_lookup(u32 index, struct tcf_hashinfo *hinfo)  { -	struct tcf_common *p; +	struct tcf_common *p = NULL; +	struct hlist_head *head; -	read_lock_bh(hinfo->lock); -	for (p = hinfo->htab[tcf_hash(index, hinfo->hmask)]; p; -	     p = p->tcfc_next) { +	spin_lock_bh(&hinfo->lock); +	head = &hinfo->htab[tcf_hash(index, hinfo->hmask)]; +	hlist_for_each_entry_rcu(p, head, tcfc_head)  		if (p->tcfc_index == index)  			break; -	} -	read_unlock_bh(hinfo->lock); +	spin_unlock_bh(&hinfo->lock);  	return p;  } -EXPORT_SYMBOL(tcf_hash_lookup); -u32 tcf_hash_new_index(u32 *idx_gen, struct tcf_hashinfo *hinfo) +u32 tcf_hash_new_index(struct tcf_hashinfo *hinfo)  { -	u32 val = *idx_gen; +	u32 val = hinfo->index;  	do {  		if (++val == 0)  			val = 1;  	} while (tcf_hash_lookup(val, hinfo)); -	return (*idx_gen = val); +	hinfo->index = val; +	return val;  }  EXPORT_SYMBOL(tcf_hash_new_index); @@ -210,34 +209,46 @@ int tcf_hash_search(struct tc_action *a, u32 index)  }  EXPORT_SYMBOL(tcf_hash_search); -struct tcf_common *tcf_hash_check(u32 index, struct tc_action *a, int bind, -				  struct tcf_hashinfo *hinfo) +int tcf_hash_check(u32 index, struct tc_action *a, int bind)  { +	struct tcf_hashinfo *hinfo = a->ops->hinfo;  	struct tcf_common *p = NULL;  	if (index && (p = tcf_hash_lookup(index, hinfo)) != NULL) {  		if (bind)  			p->tcfc_bindcnt++;  		p->tcfc_refcnt++;  		a->priv = p; +		return 1;  	} -	return p; +	return 0;  }  EXPORT_SYMBOL(tcf_hash_check); -struct tcf_common *tcf_hash_create(u32 index, struct nlattr *est, -				   struct tc_action *a, int size, int bind, -				   u32 *idx_gen, struct tcf_hashinfo *hinfo) +void tcf_hash_cleanup(struct tc_action *a, struct nlattr *est) +{ +	struct tcf_common *pc = a->priv; +	if (est) +		gen_kill_estimator(&pc->tcfc_bstats, +				   &pc->tcfc_rate_est); +	kfree_rcu(pc, tcfc_rcu); +} +EXPORT_SYMBOL(tcf_hash_cleanup); + +int tcf_hash_create(u32 index, struct nlattr *est, struct tc_action *a, +		    int size, int bind)  { +	struct tcf_hashinfo *hinfo = a->ops->hinfo;  	struct tcf_common *p = kzalloc(size, GFP_KERNEL);  	if (unlikely(!p)) -		return ERR_PTR(-ENOMEM); +		return -ENOMEM;  	p->tcfc_refcnt = 1;  	if (bind)  		p->tcfc_bindcnt = 1;  	spin_lock_init(&p->tcfc_lock); -	p->tcfc_index = index ? index : tcf_hash_new_index(idx_gen, hinfo); +	INIT_HLIST_NODE(&p->tcfc_head); +	p->tcfc_index = index ? index : tcf_hash_new_index(hinfo);  	p->tcfc_tm.install = jiffies;  	p->tcfc_tm.lastuse = jiffies;  	if (est) { @@ -245,42 +256,64 @@ struct tcf_common *tcf_hash_create(u32 index, struct nlattr *est,  					    &p->tcfc_lock, est);  		if (err) {  			kfree(p); -			return ERR_PTR(err); +			return err;  		}  	}  	a->priv = (void *) p; -	return p; +	return 0;  }  EXPORT_SYMBOL(tcf_hash_create); -void tcf_hash_insert(struct tcf_common *p, struct tcf_hashinfo *hinfo) +void tcf_hash_insert(struct tc_action *a)  { +	struct tcf_common *p = a->priv; +	struct tcf_hashinfo *hinfo = a->ops->hinfo;  	unsigned int h = tcf_hash(p->tcfc_index, hinfo->hmask); -	write_lock_bh(hinfo->lock); -	p->tcfc_next = hinfo->htab[h]; -	hinfo->htab[h] = p; -	write_unlock_bh(hinfo->lock); +	spin_lock_bh(&hinfo->lock); +	hlist_add_head(&p->tcfc_head, &hinfo->htab[h]); +	spin_unlock_bh(&hinfo->lock);  }  EXPORT_SYMBOL(tcf_hash_insert); -static struct tc_action_ops *act_base = NULL; +static LIST_HEAD(act_base);  static DEFINE_RWLOCK(act_mod_lock); -int tcf_register_action(struct tc_action_ops *act) +int tcf_register_action(struct tc_action_ops *act, unsigned int mask)  { -	struct tc_action_ops *a, **ap; +	struct tc_action_ops *a; +	int err; + +	/* Must supply act, dump and init */ +	if (!act->act || !act->dump || !act->init) +		return -EINVAL; + +	/* Supply defaults */ +	if (!act->lookup) +		act->lookup = tcf_hash_search; +	if (!act->walk) +		act->walk = tcf_generic_walker; + +	act->hinfo = kmalloc(sizeof(struct tcf_hashinfo), GFP_KERNEL); +	if (!act->hinfo) +		return -ENOMEM; +	err = tcf_hashinfo_init(act->hinfo, mask); +	if (err) { +		kfree(act->hinfo); +		return err; +	}  	write_lock(&act_mod_lock); -	for (ap = &act_base; (a = *ap) != NULL; ap = &a->next) { +	list_for_each_entry(a, &act_base, head) {  		if (act->type == a->type || (strcmp(act->kind, a->kind) == 0)) {  			write_unlock(&act_mod_lock); +			tcf_hashinfo_destroy(act->hinfo); +			kfree(act->hinfo);  			return -EEXIST;  		}  	} -	act->next = NULL; -	*ap = act; +	list_add_tail(&act->head, &act_base);  	write_unlock(&act_mod_lock);  	return 0;  } @@ -288,17 +321,18 @@ EXPORT_SYMBOL(tcf_register_action);  int tcf_unregister_action(struct tc_action_ops *act)  { -	struct tc_action_ops *a, **ap; +	struct tc_action_ops *a;  	int err = -ENOENT;  	write_lock(&act_mod_lock); -	for (ap = &act_base; (a = *ap) != NULL; ap = &a->next) -		if (a == act) +	list_for_each_entry(a, &act_base, head) { +		if (a == act) { +			list_del(&act->head); +			tcf_hashinfo_destroy(act->hinfo); +			kfree(act->hinfo); +			err = 0;  			break; -	if (a) { -		*ap = a->next; -		a->next = NULL; -		err = 0; +		}  	}  	write_unlock(&act_mod_lock);  	return err; @@ -308,72 +342,45 @@ EXPORT_SYMBOL(tcf_unregister_action);  /* lookup by name */  static struct tc_action_ops *tc_lookup_action_n(char *kind)  { -	struct tc_action_ops *a = NULL; +	struct tc_action_ops *a, *res = NULL;  	if (kind) {  		read_lock(&act_mod_lock); -		for (a = act_base; a; a = a->next) { +		list_for_each_entry(a, &act_base, head) {  			if (strcmp(kind, a->kind) == 0) { -				if (!try_module_get(a->owner)) { -					read_unlock(&act_mod_lock); -					return NULL; -				} +				if (try_module_get(a->owner)) +					res = a;  				break;  			}  		}  		read_unlock(&act_mod_lock);  	} -	return a; +	return res;  }  /* lookup by nlattr */  static struct tc_action_ops *tc_lookup_action(struct nlattr *kind)  { -	struct tc_action_ops *a = NULL; +	struct tc_action_ops *a, *res = NULL;  	if (kind) {  		read_lock(&act_mod_lock); -		for (a = act_base; a; a = a->next) { +		list_for_each_entry(a, &act_base, head) {  			if (nla_strcmp(kind, a->kind) == 0) { -				if (!try_module_get(a->owner)) { -					read_unlock(&act_mod_lock); -					return NULL; -				} +				if (try_module_get(a->owner)) +					res = a;  				break;  			}  		}  		read_unlock(&act_mod_lock);  	} -	return a; +	return res;  } -#if 0 -/* lookup by id */ -static struct tc_action_ops *tc_lookup_action_id(u32 type) -{ -	struct tc_action_ops *a = NULL; - -	if (type) { -		read_lock(&act_mod_lock); -		for (a = act_base; a; a = a->next) { -			if (a->type == type) { -				if (!try_module_get(a->owner)) { -					read_unlock(&act_mod_lock); -					return NULL; -				} -				break; -			} -		} -		read_unlock(&act_mod_lock); -	} -	return a; -} -#endif - -int tcf_action_exec(struct sk_buff *skb, struct tc_action *act, +int tcf_action_exec(struct sk_buff *skb, const struct list_head *actions,  		    struct tcf_result *res)  { -	struct tc_action *a; +	const struct tc_action *a;  	int ret = -1;  	if (skb->tc_verd & TC_NCLS) { @@ -381,53 +388,44 @@ int tcf_action_exec(struct sk_buff *skb, struct tc_action *act,  		ret = TC_ACT_OK;  		goto exec_done;  	} -	while ((a = act) != NULL) { +	list_for_each_entry(a, actions, list) {  repeat: -		if (a->ops && a->ops->act) { -			ret = a->ops->act(skb, a, res); -			if (TC_MUNGED & skb->tc_verd) { -				/* copied already, allow trampling */ -				skb->tc_verd = SET_TC_OK2MUNGE(skb->tc_verd); -				skb->tc_verd = CLR_TC_MUNGED(skb->tc_verd); -			} -			if (ret == TC_ACT_REPEAT) -				goto repeat;	/* we need a ttl - JHS */ -			if (ret != TC_ACT_PIPE) -				goto exec_done; +		ret = a->ops->act(skb, a, res); +		if (TC_MUNGED & skb->tc_verd) { +			/* copied already, allow trampling */ +			skb->tc_verd = SET_TC_OK2MUNGE(skb->tc_verd); +			skb->tc_verd = CLR_TC_MUNGED(skb->tc_verd);  		} -		act = a->next; +		if (ret == TC_ACT_REPEAT) +			goto repeat;	/* we need a ttl - JHS */ +		if (ret != TC_ACT_PIPE) +			goto exec_done;  	}  exec_done:  	return ret;  }  EXPORT_SYMBOL(tcf_action_exec); -void tcf_action_destroy(struct tc_action *act, int bind) +int tcf_action_destroy(struct list_head *actions, int bind)  { -	struct tc_action *a; +	struct tc_action *a, *tmp; +	int ret = 0; -	for (a = act; a; a = act) { -		if (a->ops && a->ops->cleanup) { -			if (a->ops->cleanup(a, bind) == ACT_P_DELETED) -				module_put(a->ops->owner); -			act = act->next; -			kfree(a); -		} else { -			/*FIXME: Remove later - catch insertion bugs*/ -			WARN(1, "tcf_action_destroy: BUG? destroying NULL ops\n"); -			act = act->next; -			kfree(a); -		} +	list_for_each_entry_safe(a, tmp, actions, list) { +		ret = tcf_hash_release(a, bind); +		if (ret == ACT_P_DELETED) +			module_put(a->ops->owner); +		else if (ret < 0) +			return ret; +		list_del(&a->list); +		kfree(a);  	} +	return ret;  }  int  tcf_action_dump_old(struct sk_buff *skb, struct tc_action *a, int bind, int ref)  { -	int err = -EINVAL; - -	if (a->ops == NULL || a->ops->dump == NULL) -		return err;  	return a->ops->dump(skb, a, bind, ref);  } @@ -438,16 +436,15 @@ tcf_action_dump_1(struct sk_buff *skb, struct tc_action *a, int bind, int ref)  	unsigned char *b = skb_tail_pointer(skb);  	struct nlattr *nest; -	if (a->ops == NULL || a->ops->dump == NULL) -		return err; - -	NLA_PUT_STRING(skb, TCA_KIND, a->ops->kind); +	if (nla_put_string(skb, TCA_KIND, a->ops->kind)) +		goto nla_put_failure;  	if (tcf_action_copy_stats(skb, a, 0))  		goto nla_put_failure;  	nest = nla_nest_start(skb, TCA_OPTIONS);  	if (nest == NULL)  		goto nla_put_failure; -	if ((err = tcf_action_dump_old(skb, a, bind, ref)) > 0) { +	err = tcf_action_dump_old(skb, a, bind, ref); +	if (err > 0) {  		nla_nest_end(skb, nest);  		return err;  	} @@ -459,14 +456,13 @@ nla_put_failure:  EXPORT_SYMBOL(tcf_action_dump_1);  int -tcf_action_dump(struct sk_buff *skb, struct tc_action *act, int bind, int ref) +tcf_action_dump(struct sk_buff *skb, struct list_head *actions, int bind, int ref)  {  	struct tc_action *a;  	int err = -EINVAL;  	struct nlattr *nest; -	while ((a = act) != NULL) { -		act = a->next; +	list_for_each_entry(a, actions, list) {  		nest = nla_nest_start(skb, a->order);  		if (nest == NULL)  			goto nla_put_failure; @@ -485,13 +481,14 @@ errout:  	return err;  } -struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est, -				    char *name, int ovr, int bind) +struct tc_action *tcf_action_init_1(struct net *net, struct nlattr *nla, +				    struct nlattr *est, char *name, int ovr, +				    int bind)  {  	struct tc_action *a;  	struct tc_action_ops *a_o;  	char act_name[IFNAMSIZ]; -	struct nlattr *tb[TCA_ACT_MAX+1]; +	struct nlattr *tb[TCA_ACT_MAX + 1];  	struct nlattr *kind;  	int err; @@ -540,21 +537,22 @@ struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est,  	if (a == NULL)  		goto err_mod; +	a->ops = a_o; +	INIT_LIST_HEAD(&a->list);  	/* backward compatibility for policer */  	if (name == NULL) -		err = a_o->init(tb[TCA_ACT_OPTIONS], est, a, ovr, bind); +		err = a_o->init(net, tb[TCA_ACT_OPTIONS], est, a, ovr, bind);  	else -		err = a_o->init(nla, est, a, ovr, bind); +		err = a_o->init(net, nla, est, a, ovr, bind);  	if (err < 0)  		goto err_free;  	/* module count goes up only when brand new policy is created -	   if it exists and is only bound to in a_o->init() then -	   ACT_P_CREATED is not returned (a zero is). -	*/ +	 * if it exists and is only bound to in a_o->init() then +	 * ACT_P_CREATED is not returned (a zero is). +	 */  	if (err != ACT_P_CREATED)  		module_put(a_o->owner); -	a->ops = a_o;  	return a; @@ -566,36 +564,33 @@ err_out:  	return ERR_PTR(err);  } -struct tc_action *tcf_action_init(struct nlattr *nla, struct nlattr *est, -				  char *name, int ovr, int bind) +int tcf_action_init(struct net *net, struct nlattr *nla, +				  struct nlattr *est, char *name, int ovr, +				  int bind, struct list_head *actions)  { -	struct nlattr *tb[TCA_ACT_MAX_PRIO+1]; -	struct tc_action *head = NULL, *act, *act_prev = NULL; +	struct nlattr *tb[TCA_ACT_MAX_PRIO + 1]; +	struct tc_action *act;  	int err;  	int i;  	err = nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL);  	if (err < 0) -		return ERR_PTR(err); +		return err;  	for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) { -		act = tcf_action_init_1(tb[i], est, name, ovr, bind); -		if (IS_ERR(act)) +		act = tcf_action_init_1(net, tb[i], est, name, ovr, bind); +		if (IS_ERR(act)) { +			err = PTR_ERR(act);  			goto err; +		}  		act->order = i; - -		if (head == NULL) -			head = act; -		else -			act_prev->next = act; -		act_prev = act; +		list_add_tail(&act->list, actions);  	} -	return head; +	return 0;  err: -	if (head != NULL) -		tcf_action_destroy(head, bind); -	return act; +	tcf_action_destroy(actions, bind); +	return err;  }  int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a, @@ -603,9 +598,9 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,  {  	int err = 0;  	struct gnet_dump d; -	struct tcf_act_hdr *h = a->priv; +	struct tcf_common *p = a->priv; -	if (h == NULL) +	if (p == NULL)  		goto errout;  	/* compat_mode being true specifies a call that is supposed @@ -614,24 +609,20 @@ int tcf_action_copy_stats(struct sk_buff *skb, struct tc_action *a,  	if (compat_mode) {  		if (a->type == TCA_OLD_COMPAT)  			err = gnet_stats_start_copy_compat(skb, 0, -				TCA_STATS, TCA_XSTATS, &h->tcf_lock, &d); +				TCA_STATS, TCA_XSTATS, &p->tcfc_lock, &d);  		else  			return 0;  	} else  		err = gnet_stats_start_copy(skb, TCA_ACT_STATS, -					    &h->tcf_lock, &d); +					    &p->tcfc_lock, &d);  	if (err < 0)  		goto errout; -	if (a->ops != NULL && a->ops->get_stats != NULL) -		if (a->ops->get_stats(skb, a) < 0) -			goto errout; - -	if (gnet_stats_copy_basic(&d, &h->tcf_bstats) < 0 || -	    gnet_stats_copy_rate_est(&d, &h->tcf_bstats, -				     &h->tcf_rate_est) < 0 || -	    gnet_stats_copy_queue(&d, &h->tcf_qstats) < 0) +	if (gnet_stats_copy_basic(&d, &p->tcfc_bstats) < 0 || +	    gnet_stats_copy_rate_est(&d, &p->tcfc_bstats, +				     &p->tcfc_rate_est) < 0 || +	    gnet_stats_copy_queue(&d, &p->tcfc_qstats) < 0)  		goto errout;  	if (gnet_stats_finish_copy(&d) < 0) @@ -644,7 +635,7 @@ errout:  }  static int -tca_get_fill(struct sk_buff *skb, struct tc_action *a, u32 pid, u32 seq, +tca_get_fill(struct sk_buff *skb, struct list_head *actions, u32 portid, u32 seq,  	     u16 flags, int event, int bind, int ref)  {  	struct tcamsg *t; @@ -652,52 +643,66 @@ tca_get_fill(struct sk_buff *skb, struct tc_action *a, u32 pid, u32 seq,  	unsigned char *b = skb_tail_pointer(skb);  	struct nlattr *nest; -	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*t), flags); - -	t = NLMSG_DATA(nlh); +	nlh = nlmsg_put(skb, portid, seq, event, sizeof(*t), flags); +	if (!nlh) +		goto out_nlmsg_trim; +	t = nlmsg_data(nlh);  	t->tca_family = AF_UNSPEC;  	t->tca__pad1 = 0;  	t->tca__pad2 = 0;  	nest = nla_nest_start(skb, TCA_ACT_TAB);  	if (nest == NULL) -		goto nla_put_failure; +		goto out_nlmsg_trim; -	if (tcf_action_dump(skb, a, bind, ref) < 0) -		goto nla_put_failure; +	if (tcf_action_dump(skb, actions, bind, ref) < 0) +		goto out_nlmsg_trim;  	nla_nest_end(skb, nest);  	nlh->nlmsg_len = skb_tail_pointer(skb) - b;  	return skb->len; -nla_put_failure: -nlmsg_failure: +out_nlmsg_trim:  	nlmsg_trim(skb, b);  	return -1;  }  static int -act_get_notify(struct net *net, u32 pid, struct nlmsghdr *n, -	       struct tc_action *a, int event) +act_get_notify(struct net *net, u32 portid, struct nlmsghdr *n, +	       struct list_head *actions, int event)  {  	struct sk_buff *skb;  	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);  	if (!skb)  		return -ENOBUFS; -	if (tca_get_fill(skb, a, pid, n->nlmsg_seq, 0, event, 0, 0) <= 0) { +	if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, event, 0, 0) <= 0) {  		kfree_skb(skb);  		return -EINVAL;  	} -	return rtnl_unicast(skb, net, pid); +	return rtnl_unicast(skb, net, portid); +} + +static struct tc_action *create_a(int i) +{ +	struct tc_action *act; + +	act = kzalloc(sizeof(*act), GFP_KERNEL); +	if (act == NULL) { +		pr_debug("create_a: failed to alloc!\n"); +		return NULL; +	} +	act->order = i; +	INIT_LIST_HEAD(&act->list); +	return act;  }  static struct tc_action * -tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 pid) +tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 portid)  { -	struct nlattr *tb[TCA_ACT_MAX+1]; +	struct nlattr *tb[TCA_ACT_MAX + 1];  	struct tc_action *a;  	int index;  	int err; @@ -713,16 +718,14 @@ tcf_action_get_1(struct nlattr *nla, struct nlmsghdr *n, u32 pid)  	index = nla_get_u32(tb[TCA_ACT_INDEX]);  	err = -ENOMEM; -	a = kzalloc(sizeof(struct tc_action), GFP_KERNEL); +	a = create_a(0);  	if (a == NULL)  		goto err_out;  	err = -EINVAL;  	a->ops = tc_lookup_action(tb[TCA_ACT_KIND]); -	if (a->ops == NULL) +	if (a->ops == NULL) /* could happen in batch of actions */  		goto err_free; -	if (a->ops->lookup == NULL) -		goto err_mod;  	err = -ENOENT;  	if (a->ops->lookup(a, index) == 0)  		goto err_mod; @@ -738,31 +741,18 @@ err_out:  	return ERR_PTR(err);  } -static void cleanup_a(struct tc_action *act) +static void cleanup_a(struct list_head *actions)  { -	struct tc_action *a; +	struct tc_action *a, *tmp; -	for (a = act; a; a = act) { -		act = a->next; +	list_for_each_entry_safe(a, tmp, actions, list) { +		list_del(&a->list);  		kfree(a);  	}  } -static struct tc_action *create_a(int i) -{ -	struct tc_action *act; - -	act = kzalloc(sizeof(*act), GFP_KERNEL); -	if (act == NULL) { -		pr_debug("create_a: failed to alloc!\n"); -		return NULL; -	} -	act->order = i; -	return act; -} -  static int tca_action_flush(struct net *net, struct nlattr *nla, -			    struct nlmsghdr *n, u32 pid) +			    struct nlmsghdr *n, u32 portid)  {  	struct sk_buff *skb;  	unsigned char *b; @@ -770,20 +760,14 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,  	struct tcamsg *t;  	struct netlink_callback dcb;  	struct nlattr *nest; -	struct nlattr *tb[TCA_ACT_MAX+1]; +	struct nlattr *tb[TCA_ACT_MAX + 1];  	struct nlattr *kind; -	struct tc_action *a = create_a(0); +	struct tc_action a;  	int err = -ENOMEM; -	if (a == NULL) { -		pr_debug("tca_action_flush: couldnt create tc_action\n"); -		return err; -	} -  	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);  	if (!skb) {  		pr_debug("tca_action_flush: failed skb alloc\n"); -		kfree(a);  		return err;  	} @@ -795,23 +779,27 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,  	err = -EINVAL;  	kind = tb[TCA_ACT_KIND]; -	a->ops = tc_lookup_action(kind); -	if (a->ops == NULL) +	memset(&a, 0, sizeof(struct tc_action)); +	INIT_LIST_HEAD(&a.list); +	a.ops = tc_lookup_action(kind); +	if (a.ops == NULL) /*some idjot trying to flush unknown action */  		goto err_out; -	nlh = NLMSG_PUT(skb, pid, n->nlmsg_seq, RTM_DELACTION, sizeof(*t)); -	t = NLMSG_DATA(nlh); +	nlh = nlmsg_put(skb, portid, n->nlmsg_seq, RTM_DELACTION, sizeof(*t), 0); +	if (!nlh) +		goto out_module_put; +	t = nlmsg_data(nlh);  	t->tca_family = AF_UNSPEC;  	t->tca__pad1 = 0;  	t->tca__pad2 = 0;  	nest = nla_nest_start(skb, TCA_ACT_TAB);  	if (nest == NULL) -		goto nla_put_failure; +		goto out_module_put; -	err = a->ops->walk(skb, &dcb, RTM_DELACTION, a); +	err = a.ops->walk(skb, &dcb, RTM_DELACTION, &a);  	if (err < 0) -		goto nla_put_failure; +		goto out_module_put;  	if (err == 0)  		goto noflush_out; @@ -819,171 +807,150 @@ static int tca_action_flush(struct net *net, struct nlattr *nla,  	nlh->nlmsg_len = skb_tail_pointer(skb) - b;  	nlh->nlmsg_flags |= NLM_F_ROOT; -	module_put(a->ops->owner); -	kfree(a); -	err = rtnetlink_send(skb, net, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO); +	module_put(a.ops->owner); +	err = rtnetlink_send(skb, net, portid, RTNLGRP_TC, +			     n->nlmsg_flags & NLM_F_ECHO);  	if (err > 0)  		return 0;  	return err; -nla_put_failure: -nlmsg_failure: -	module_put(a->ops->owner); +out_module_put: +	module_put(a.ops->owner);  err_out:  noflush_out:  	kfree_skb(skb); -	kfree(a);  	return err;  }  static int +tcf_del_notify(struct net *net, struct nlmsghdr *n, struct list_head *actions, +	       u32 portid) +{ +	int ret; +	struct sk_buff *skb; + +	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); +	if (!skb) +		return -ENOBUFS; + +	if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, 0, RTM_DELACTION, +			 0, 1) <= 0) { +		kfree_skb(skb); +		return -EINVAL; +	} + +	/* now do the delete */ +	ret = tcf_action_destroy(actions, 0); +	if (ret < 0) { +		kfree_skb(skb); +		return ret; +	} + +	ret = rtnetlink_send(skb, net, portid, RTNLGRP_TC, +			     n->nlmsg_flags & NLM_F_ECHO); +	if (ret > 0) +		return 0; +	return ret; +} + +static int  tca_action_gd(struct net *net, struct nlattr *nla, struct nlmsghdr *n, -	      u32 pid, int event) +	      u32 portid, int event)  {  	int i, ret; -	struct nlattr *tb[TCA_ACT_MAX_PRIO+1]; -	struct tc_action *head = NULL, *act, *act_prev = NULL; +	struct nlattr *tb[TCA_ACT_MAX_PRIO + 1]; +	struct tc_action *act; +	LIST_HEAD(actions);  	ret = nla_parse_nested(tb, TCA_ACT_MAX_PRIO, nla, NULL);  	if (ret < 0)  		return ret; -	if (event == RTM_DELACTION && n->nlmsg_flags&NLM_F_ROOT) { +	if (event == RTM_DELACTION && n->nlmsg_flags & NLM_F_ROOT) {  		if (tb[1] != NULL) -			return tca_action_flush(net, tb[1], n, pid); +			return tca_action_flush(net, tb[1], n, portid);  		else  			return -EINVAL;  	}  	for (i = 1; i <= TCA_ACT_MAX_PRIO && tb[i]; i++) { -		act = tcf_action_get_1(tb[i], n, pid); +		act = tcf_action_get_1(tb[i], n, portid);  		if (IS_ERR(act)) {  			ret = PTR_ERR(act);  			goto err;  		}  		act->order = i; - -		if (head == NULL) -			head = act; -		else -			act_prev->next = act; -		act_prev = act; +		list_add_tail(&act->list, &actions);  	}  	if (event == RTM_GETACTION) -		ret = act_get_notify(net, pid, n, head, event); +		ret = act_get_notify(net, portid, n, &actions, event);  	else { /* delete */ -		struct sk_buff *skb; - -		skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); -		if (!skb) { -			ret = -ENOBUFS; -			goto err; -		} - -		if (tca_get_fill(skb, head, pid, n->nlmsg_seq, 0, event, -				 0, 1) <= 0) { -			kfree_skb(skb); -			ret = -EINVAL; +		ret = tcf_del_notify(net, n, &actions, portid); +		if (ret)  			goto err; -		} - -		/* now do the delete */ -		tcf_action_destroy(head, 0); -		ret = rtnetlink_send(skb, net, pid, RTNLGRP_TC, -				     n->nlmsg_flags&NLM_F_ECHO); -		if (ret > 0) -			return 0;  		return ret;  	}  err: -	cleanup_a(head); +	cleanup_a(&actions);  	return ret;  } -static int tcf_add_notify(struct net *net, struct tc_action *a, -			  u32 pid, u32 seq, int event, u16 flags) +static int +tcf_add_notify(struct net *net, struct nlmsghdr *n, struct list_head *actions, +	       u32 portid)  { -	struct tcamsg *t; -	struct nlmsghdr *nlh;  	struct sk_buff *skb; -	struct nlattr *nest; -	unsigned char *b;  	int err = 0;  	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);  	if (!skb)  		return -ENOBUFS; -	b = skb_tail_pointer(skb); - -	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*t), flags); -	t = NLMSG_DATA(nlh); -	t->tca_family = AF_UNSPEC; -	t->tca__pad1 = 0; -	t->tca__pad2 = 0; - -	nest = nla_nest_start(skb, TCA_ACT_TAB); -	if (nest == NULL) -		goto nla_put_failure; - -	if (tcf_action_dump(skb, a, 0, 0) < 0) -		goto nla_put_failure; - -	nla_nest_end(skb, nest); - -	nlh->nlmsg_len = skb_tail_pointer(skb) - b; -	NETLINK_CB(skb).dst_group = RTNLGRP_TC; +	if (tca_get_fill(skb, actions, portid, n->nlmsg_seq, n->nlmsg_flags, +			 RTM_NEWACTION, 0, 0) <= 0) { +		kfree_skb(skb); +		return -EINVAL; +	} -	err = rtnetlink_send(skb, net, pid, RTNLGRP_TC, flags&NLM_F_ECHO); +	err = rtnetlink_send(skb, net, portid, RTNLGRP_TC, +			     n->nlmsg_flags & NLM_F_ECHO);  	if (err > 0)  		err = 0;  	return err; - -nla_put_failure: -nlmsg_failure: -	kfree_skb(skb); -	return -1;  } -  static int  tcf_action_add(struct net *net, struct nlattr *nla, struct nlmsghdr *n, -	       u32 pid, int ovr) +	       u32 portid, int ovr)  {  	int ret = 0; -	struct tc_action *act; -	struct tc_action *a; -	u32 seq = n->nlmsg_seq; +	LIST_HEAD(actions); -	act = tcf_action_init(nla, NULL, NULL, ovr, 0); -	if (act == NULL) -		goto done; -	if (IS_ERR(act)) { -		ret = PTR_ERR(act); +	ret = tcf_action_init(net, nla, NULL, NULL, ovr, 0, &actions); +	if (ret)  		goto done; -	}  	/* dump then free all the actions after update; inserted policy  	 * stays intact -	 * */ -	ret = tcf_add_notify(net, act, pid, seq, RTM_NEWACTION, n->nlmsg_flags); -	for (a = act; a; a = act) { -		act = a->next; -		kfree(a); -	} +	 */ +	ret = tcf_add_notify(net, n, &actions, portid); +	cleanup_a(&actions);  done:  	return ret;  } -static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, void *arg) +static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n)  {  	struct net *net = sock_net(skb->sk);  	struct nlattr *tca[TCA_ACT_MAX + 1]; -	u32 pid = skb ? NETLINK_CB(skb).pid : 0; +	u32 portid = skb ? NETLINK_CB(skb).portid : 0;  	int ret = 0, ovr = 0; +	if ((n->nlmsg_type != RTM_GETACTION) && !netlink_capable(skb, CAP_NET_ADMIN)) +		return -EPERM; +  	ret = nlmsg_parse(n, sizeof(struct tcamsg), tca, TCA_ACT_MAX, NULL);  	if (ret < 0)  		return ret; @@ -993,30 +960,29 @@ static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, void *arg)  		return -EINVAL;  	} -	/* n->nlmsg_flags&NLM_F_CREATE -	 * */ +	/* n->nlmsg_flags & NLM_F_CREATE */  	switch (n->nlmsg_type) {  	case RTM_NEWACTION:  		/* we are going to assume all other flags -		 * imply create only if it doesnt exist +		 * imply create only if it doesn't exist  		 * Note that CREATE | EXCL implies that  		 * but since we want avoid ambiguity (eg when flags  		 * is zero) then just set this  		 */ -		if (n->nlmsg_flags&NLM_F_REPLACE) +		if (n->nlmsg_flags & NLM_F_REPLACE)  			ovr = 1;  replay: -		ret = tcf_action_add(net, tca[TCA_ACT_TAB], n, pid, ovr); +		ret = tcf_action_add(net, tca[TCA_ACT_TAB], n, portid, ovr);  		if (ret == -EAGAIN)  			goto replay;  		break;  	case RTM_DELACTION:  		ret = tca_action_gd(net, tca[TCA_ACT_TAB], n, -				    pid, RTM_DELACTION); +				    portid, RTM_DELACTION);  		break;  	case RTM_GETACTION:  		ret = tca_action_gd(net, tca[TCA_ACT_TAB], n, -				    pid, RTM_GETACTION); +				    portid, RTM_GETACTION);  		break;  	default:  		BUG(); @@ -1028,7 +994,7 @@ replay:  static struct nlattr *  find_dump_kind(const struct nlmsghdr *n)  { -	struct nlattr *tb1, *tb2[TCA_ACT_MAX+1]; +	struct nlattr *tb1, *tb2[TCA_ACT_MAX + 1];  	struct nlattr *tb[TCA_ACT_MAX_PRIO + 1];  	struct nlattr *nla[TCAA_MAX + 1];  	struct nlattr *kind; @@ -1062,7 +1028,7 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)  	struct tc_action_ops *a_o;  	struct tc_action a;  	int ret = 0; -	struct tcamsg *t = (struct tcamsg *) NLMSG_DATA(cb->nlh); +	struct tcamsg *t = (struct tcamsg *) nlmsg_data(cb->nlh);  	struct nlattr *kind = find_dump_kind(cb->nlh);  	if (kind == NULL) { @@ -1071,33 +1037,28 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)  	}  	a_o = tc_lookup_action(kind); -	if (a_o == NULL) { +	if (a_o == NULL)  		return 0; -	}  	memset(&a, 0, sizeof(struct tc_action));  	a.ops = a_o; -	if (a_o->walk == NULL) { -		WARN(1, "tc_dump_action: %s !capable of dumping table\n", -		     a_o->kind); -		goto nla_put_failure; -	} - -	nlh = NLMSG_PUT(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, -			cb->nlh->nlmsg_type, sizeof(*t)); -	t = NLMSG_DATA(nlh); +	nlh = nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, +			cb->nlh->nlmsg_type, sizeof(*t), 0); +	if (!nlh) +		goto out_module_put; +	t = nlmsg_data(nlh);  	t->tca_family = AF_UNSPEC;  	t->tca__pad1 = 0;  	t->tca__pad2 = 0;  	nest = nla_nest_start(skb, TCA_ACT_TAB);  	if (nest == NULL) -		goto nla_put_failure; +		goto out_module_put;  	ret = a_o->walk(skb, cb, RTM_GETACTION, &a);  	if (ret < 0) -		goto nla_put_failure; +		goto out_module_put;  	if (ret > 0) {  		nla_nest_end(skb, nest); @@ -1106,13 +1067,12 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb)  		nla_nest_cancel(skb, nest);  	nlh->nlmsg_len = skb_tail_pointer(skb) - b; -	if (NETLINK_CB(cb->skb).pid && ret) +	if (NETLINK_CB(cb->skb).portid && ret)  		nlh->nlmsg_flags |= NLM_F_MULTI;  	module_put(a_o->owner);  	return skb->len; -nla_put_failure: -nlmsg_failure: +out_module_put:  	module_put(a_o->owner);  	nlmsg_trim(skb, b);  	return skb->len; @@ -1120,9 +1080,10 @@ nlmsg_failure:  static int __init tc_action_init(void)  { -	rtnl_register(PF_UNSPEC, RTM_NEWACTION, tc_ctl_action, NULL); -	rtnl_register(PF_UNSPEC, RTM_DELACTION, tc_ctl_action, NULL); -	rtnl_register(PF_UNSPEC, RTM_GETACTION, tc_ctl_action, tc_dump_action); +	rtnl_register(PF_UNSPEC, RTM_NEWACTION, tc_ctl_action, NULL, NULL); +	rtnl_register(PF_UNSPEC, RTM_DELACTION, tc_ctl_action, NULL, NULL); +	rtnl_register(PF_UNSPEC, RTM_GETACTION, tc_ctl_action, tc_dump_action, +		      NULL);  	return 0;  }  | 
