fib: cleanups (6a31d2a9) · Commits · e / devices / android_kernel_xiaomi_markw

net/ipv4/fib_frontend.c

+66 −55

Original line number	Diff line number	Diff line
		@@ -230,25 +230,28 @@ unsigned int inet_dev_addr_type(struct net net, const struct net_device dev,
		EXPORT_SYMBOL(inet_dev_addr_type);

		/* Given (packet source, input interface) and optional (dst, oif, tos):
		- (main) check, that source is valid i.e. not broadcast or our local
		address.
		- figure out what "logical" interface this packet arrived
		and calculate "specific destination" address.
		- check, that packet arrived from expected physical interface.
		* - (main) check, that source is valid i.e. not broadcast or our local
		* address.
		* - figure out what "logical" interface this packet arrived
		* and calculate "specific destination" address.
		* - check, that packet arrived from expected physical interface.
		*/

		int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif,
		struct net_device dev, __be32 spec_dst,
		u32 *itag, u32 mark)
		{
		struct in_device *in_dev;
		struct flowi fl = { .nl_u = { .ip4_u =
		{ .daddr = src,
		struct flowi fl = {
		.nl_u = {
		.ip4_u = {
		.daddr = src,
		.saddr = dst,
		.tos = tos } },
		.tos = tos
		}
		},
		.mark = mark,
		.iif = oif };

		.iif = oif
		};
		struct fib_result res;
		int no_addr, rpf, accept_local;
		bool dev_match;
		@@ -477,9 +480,9 @@ static int rtentry_to_fib_config(struct net net, int cmd, struct rtentry rt,
		}

		/*
		* Handle IP routing ioctl calls. These are used to manipulate the routing tables
		* Handle IP routing ioctl calls.
		* These are used to manipulate the routing tables
		*/

		int ip_rt_ioctl(struct net net, unsigned int cmd, void __user arg)
		{
		struct fib_config cfg;
		@@ -692,12 +695,11 @@ out:
		}

		/* Prepare and feed intra-kernel routing request.
		Really, it should be netlink message, but :-( netlink
		can be not configured, so that we feed it directly
		to fib engine. It is legal, because all events occur
		only when netlink is already locked.
		* Really, it should be netlink message, but :-( netlink
		* can be not configured, so that we feed it directly
		* to fib engine. It is legal, because all events occur
		* only when netlink is already locked.
		*/

		static void fib_magic(int cmd, int type, __be32 dst, int dst_len, struct in_ifaddr *ifa)
		{
		struct net *net = dev_net(ifa->ifa_dev->dev);
		@@ -764,13 +766,15 @@ void fib_add_ifaddr(struct in_ifaddr *ifa)

		if (!ipv4_is_zeronet(prefix) && !(ifa->ifa_flags & IFA_F_SECONDARY) &&
		(prefix != addr \|\| ifa->ifa_prefixlen < 32)) {
		fib_magic(RTM_NEWROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL :
		RTN_UNICAST, prefix, ifa->ifa_prefixlen, prim);
		fib_magic(RTM_NEWROUTE,
		dev->flags & IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST,
		prefix, ifa->ifa_prefixlen, prim);

		/* Add network specific broadcasts, when it takes a sense */
		if (ifa->ifa_prefixlen < 31) {
		fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix, 32, prim);
		fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix\|~mask, 32, prim);
		fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix \| ~mask,
		32, prim);
		}
		}
		}
		@@ -790,8 +794,9 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
		unsigned ok = 0;

		if (!(ifa->ifa_flags & IFA_F_SECONDARY))
		fib_magic(RTM_DELROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL :
		RTN_UNICAST, any, ifa->ifa_prefixlen, prim);
		fib_magic(RTM_DELROUTE,
		dev->flags & IFF_LOOPBACK ? RTN_LOCAL : RTN_UNICAST,
		any, ifa->ifa_prefixlen, prim);
		else {
		prim = inet_ifa_byprefix(in_dev, any, ifa->ifa_mask);
		if (prim == NULL) {
		@@ -801,9 +806,9 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
		}

		/* Deletion is more complicated than add.
		We should take care of not to delete too much :-)

		Scan address list to be sure that addresses are really gone.
		* We should take care of not to delete too much :-)
		*
		* Scan address list to be sure that addresses are really gone.
		*/

		for (ifa1 = in_dev->ifa_list; ifa1; ifa1 = ifa1->ifa_next) {
		@@ -829,10 +834,10 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa)
		/* Check, that this local address finally disappeared. */
		if (inet_addr_type(dev_net(dev), ifa->ifa_local) != RTN_LOCAL) {
		/* And the last, but not the least thing.
		We must flush stray FIB entries.

		First of all, we scan fib_info list searching
		for stray nexthop entries, then ignite fib_flush.
		* We must flush stray FIB entries.
		*
		* First of all, we scan fib_info list searching
		* for stray nexthop entries, then ignite fib_flush.
		*/
		if (fib_sync_down_addr(dev_net(dev), ifa->ifa_local))
		fib_flush(dev_net(dev));
		@@ -848,10 +853,16 @@ static void nl_fib_lookup(struct fib_result_nl frn, struct fib_table tb )
		{

		struct fib_result res;
		struct flowi fl = { .mark = frn->fl_mark,
		.nl_u = { .ip4_u = { .daddr = frn->fl_addr,
		struct flowi fl = {
		.mark = frn->fl_mark,
		.nl_u = {
		.ip4_u = {
		.daddr = frn->fl_addr,
		.tos = frn->fl_tos,
		.scope = frn->fl_scope } } };
		.scope = frn->fl_scope
		}
		}
		};

		#ifdef CONFIG_IP_MULTIPLE_TABLES
		res.r = NULL;
		@@ -947,7 +958,7 @@ static int fib_inetaddr_event(struct notifier_block *this, unsigned long event,
		fib_del_ifaddr(ifa);
		if (ifa->ifa_dev->ifa_list == NULL) {
		/* Last address was deleted from this interface.
		Disable IP.
		* Disable IP.
		*/
		fib_disable_ip(dev, 1, 0);
		} else {

net/ipv4/fib_rules.c

+5 −5

Original line number	Diff line number	Diff line
		@@ -32,8 +32,7 @@
		#include <net/ip_fib.h>
		#include <net/fib_rules.h>

		struct fib4_rule
		{
		struct fib4_rule {
		struct fib_rule common;
		u8 dst_len;
		u8 src_len;
		@@ -91,7 +90,8 @@ static int fib4_rule_action(struct fib_rule rule, struct flowi flp,
		goto errout;
		}

		if ((tbl = fib_get_table(rule->fr_net, rule->table)) == NULL)
		tbl = fib_get_table(rule->fr_net, rule->table);
		if (!tbl)
		goto errout;

		err = fib_table_lookup(tbl, flp, (struct fib_result *) arg->result);

net/ipv4/fib_semantics.c

+135 −122

Original line number	Diff line number	Diff line
		@@ -60,20 +60,29 @@ static struct hlist_head fib_info_devhash[DEVINDEX_HASHSIZE];

		static DEFINE_SPINLOCK(fib_multipath_lock);

		#define for_nexthops(fi) { int nhsel; const struct fib_nh * nh; \
		for (nhsel=0, nh = (fi)->fib_nh; nhsel < (fi)->fib_nhs; nh++, nhsel++)

		#define change_nexthops(fi) { int nhsel; struct fib_nh *nexthop_nh; \
		for (nhsel=0, nexthop_nh = (struct fib_nh *)((fi)->fib_nh); nhsel < (fi)->fib_nhs; nexthop_nh++, nhsel++)
		#define for_nexthops(fi) { \
		int nhsel; const struct fib_nh *nh; \
		for (nhsel = 0, nh = (fi)->fib_nh; \
		nhsel < (fi)->fib_nhs; \
		nh++, nhsel++)

		#define change_nexthops(fi) { \
		int nhsel; struct fib_nh *nexthop_nh; \
		for (nhsel = 0, nexthop_nh = (struct fib_nh *)((fi)->fib_nh); \
		nhsel < (fi)->fib_nhs; \
		nexthop_nh++, nhsel++)

		#else /* CONFIG_IP_ROUTE_MULTIPATH */

		/* Hope, that gcc will optimize it to get rid of dummy loop */

		#define for_nexthops(fi) { int nhsel = 0; const struct fib_nh * nh = (fi)->fib_nh; \
		#define for_nexthops(fi) { \
		int nhsel; const struct fib_nh *nh = (fi)->fib_nh; \
		for (nhsel = 0; nhsel < 1; nhsel++)

		#define change_nexthops(fi) { int nhsel = 0; struct fib_nh nexthop_nh = (struct fib_nh )((fi)->fib_nh); \
		#define change_nexthops(fi) { \
		int nhsel; \
		struct fib_nh nexthop_nh = (struct fib_nh )((fi)->fib_nh); \
		for (nhsel = 0; nhsel < 1; nhsel++)

		#endif /* CONFIG_IP_ROUTE_MULTIPATH */
		@@ -86,54 +95,54 @@ static const struct
		int error;
		u8 scope;
		} fib_props[RTN_MAX + 1] = {
		{
		[RTN_UNSPEC] = {
		.error = 0,
		.scope = RT_SCOPE_NOWHERE,
		}, /* RTN_UNSPEC */
		{
		},
		[RTN_UNICAST] = {
		.error = 0,
		.scope = RT_SCOPE_UNIVERSE,
		}, /* RTN_UNICAST */
		{
		},
		[RTN_LOCAL] = {
		.error = 0,
		.scope = RT_SCOPE_HOST,
		}, /* RTN_LOCAL */
		{
		},
		[RTN_BROADCAST] = {
		.error = 0,
		.scope = RT_SCOPE_LINK,
		}, /* RTN_BROADCAST */
		{
		},
		[RTN_ANYCAST] = {
		.error = 0,
		.scope = RT_SCOPE_LINK,
		}, /* RTN_ANYCAST */
		{
		},
		[RTN_MULTICAST] = {
		.error = 0,
		.scope = RT_SCOPE_UNIVERSE,
		}, /* RTN_MULTICAST */
		{
		},
		[RTN_BLACKHOLE] = {
		.error = -EINVAL,
		.scope = RT_SCOPE_UNIVERSE,
		}, /* RTN_BLACKHOLE */
		{
		},
		[RTN_UNREACHABLE] = {
		.error = -EHOSTUNREACH,
		.scope = RT_SCOPE_UNIVERSE,
		}, /* RTN_UNREACHABLE */
		{
		},
		[RTN_PROHIBIT] = {
		.error = -EACCES,
		.scope = RT_SCOPE_UNIVERSE,
		}, /* RTN_PROHIBIT */
		{
		},
		[RTN_THROW] = {
		.error = -EAGAIN,
		.scope = RT_SCOPE_UNIVERSE,
		}, /* RTN_THROW */
		{
		},
		[RTN_NAT] = {
		.error = -EINVAL,
		.scope = RT_SCOPE_NOWHERE,
		}, /* RTN_NAT */
		{
		},
		[RTN_XRESOLVE] = {
		.error = -EINVAL,
		.scope = RT_SCOPE_NOWHERE,
		}, /* RTN_XRESOLVE */
		},
		};


		@@ -142,7 +151,7 @@ static const struct
		void free_fib_info(struct fib_info *fi)
		{
		if (fi->fib_dead == 0) {
		printk(KERN_WARNING "Freeing alive fib_info %p\n", fi);
		pr_warning("Freeing alive fib_info %p\n", fi);
		return;
		}
		change_nexthops(fi) {
		@@ -173,7 +182,7 @@ void fib_release_info(struct fib_info *fi)
		spin_unlock_bh(&fib_info_lock);
		}

		static __inline__ int nh_comp(const struct fib_info fi, const struct fib_info ofi)
		static inline int nh_comp(const struct fib_info fi, const struct fib_info ofi)
		{
		const struct fib_nh *onh = ofi->fib_nh;

		@@ -247,9 +256,8 @@ static struct fib_info fib_find_info(const struct fib_info nfi)
		}

		/* Check, that the gateway is already configured.
		Used only by redirect accept routine.
		* Used only by redirect accept routine.
		*/

		int ip_fib_check_default(__be32 gw, struct net_device *dev)
		{
		struct hlist_head *head;
		@@ -476,67 +484,67 @@ int fib_nh_match(struct fib_config cfg, struct fib_info fi)


		/*
		Picture
		-------

		Semantics of nexthop is very messy by historical reasons.
		We have to take into account, that:
		a) gateway can be actually local interface address,
		so that gatewayed route is direct.
		b) gateway must be on-link address, possibly
		described not by an ifaddr, but also by a direct route.
		c) If both gateway and interface are specified, they should not
		contradict.
		d) If we use tunnel routes, gateway could be not on-link.

		Attempt to reconcile all of these (alas, self-contradictory) conditions
		results in pretty ugly and hairy code with obscure logic.

		I chose to generalized it instead, so that the size
		of code does not increase practically, but it becomes
		much more general.
		Every prefix is assigned a "scope" value: "host" is local address,
		"link" is direct route,
		[ ... "site" ... "interior" ... ]
		and "universe" is true gateway route with global meaning.

		Every prefix refers to a set of "nexthop"s (gw, oif),
		where gw must have narrower scope. This recursion stops
		when gw has LOCAL scope or if "nexthop" is declared ONLINK,
		which means that gw is forced to be on link.

		Code is still hairy, but now it is apparently logically
		consistent and very flexible. F.e. as by-product it allows
		to co-exists in peace independent exterior and interior
		routing processes.

		Normally it looks as following.

		{universe prefix} -> (gw, oif) [scope link]
		\|
		\|-> {link prefix} -> (gw, oif) [scope local]
		\|
		\|-> {local prefix} (terminal node)
		* Picture
		* -------
		*
		* Semantics of nexthop is very messy by historical reasons.
		* We have to take into account, that:
		* a) gateway can be actually local interface address,
		* so that gatewayed route is direct.
		* b) gateway must be on-link address, possibly
		* described not by an ifaddr, but also by a direct route.
		* c) If both gateway and interface are specified, they should not
		* contradict.
		* d) If we use tunnel routes, gateway could be not on-link.
		*
		* Attempt to reconcile all of these (alas, self-contradictory) conditions
		* results in pretty ugly and hairy code with obscure logic.
		*
		* I chose to generalized it instead, so that the size
		* of code does not increase practically, but it becomes
		* much more general.
		* Every prefix is assigned a "scope" value: "host" is local address,
		* "link" is direct route,
		* [ ... "site" ... "interior" ... ]
		* and "universe" is true gateway route with global meaning.
		*
		* Every prefix refers to a set of "nexthop"s (gw, oif),
		* where gw must have narrower scope. This recursion stops
		* when gw has LOCAL scope or if "nexthop" is declared ONLINK,
		* which means that gw is forced to be on link.
		*
		* Code is still hairy, but now it is apparently logically
		* consistent and very flexible. F.e. as by-product it allows
		* to co-exists in peace independent exterior and interior
		* routing processes.
		*
		* Normally it looks as following.
		*
		* {universe prefix} -> (gw, oif) [scope link]
		* \|
		* \|-> {link prefix} -> (gw, oif) [scope local]
		* \|
		* \|-> {local prefix} (terminal node)
		*/

		static int fib_check_nh(struct fib_config cfg, struct fib_info fi,
		struct fib_nh *nh)
		{
		int err;
		struct net *net;
		struct net_device *dev;

		net = cfg->fc_nlinfo.nl_net;
		if (nh->nh_gw) {
		struct fib_result res;

		if (nh->nh_flags & RTNH_F_ONLINK) {
		struct net_device *dev;

		if (cfg->fc_scope >= RT_SCOPE_LINK)
		return -EINVAL;
		if (inet_addr_type(net, nh->nh_gw) != RTN_UNICAST)
		return -EINVAL;
		if ((dev = __dev_get_by_index(net, nh->nh_oif)) == NULL)
		dev = __dev_get_by_index(net, nh->nh_oif);
		if (!dev)
		return -ENODEV;
		if (!(dev->flags & IFF_UP))
		return -ENETDOWN;
		@@ -559,7 +567,8 @@ static int fib_check_nh(struct fib_config cfg, struct fib_info fi,
		/* It is not necessary, but requires a bit of thinking */
		if (fl.fl4_scope < RT_SCOPE_LINK)
		fl.fl4_scope = RT_SCOPE_LINK;
		if ((err = fib_lookup(net, &fl, &res)) != 0)
		err = fib_lookup(net, &fl, &res);
		if (err)
		return err;
		}
		err = -EINVAL;
		@@ -567,11 +576,12 @@ static int fib_check_nh(struct fib_config cfg, struct fib_info fi,
		goto out;
		nh->nh_scope = res.scope;
		nh->nh_oif = FIB_RES_OIF(res);
		if ((nh->nh_dev = FIB_RES_DEV(res)) == NULL)
		nh->nh_dev = dev = FIB_RES_DEV(res);
		if (!dev)
		goto out;
		dev_hold(nh->nh_dev);
		dev_hold(dev);
		err = -ENETDOWN;
		if (!(nh->nh_dev->flags & IFF_UP))
		if (!(dev->flags & IFF_UP))
		goto out;
		err = 0;
		out:
		@@ -602,7 +612,9 @@ static inline unsigned int fib_laddr_hashfn(__be32 val)
		{
		unsigned int mask = (fib_hash_size - 1);

		return ((__force u32)val ^ ((__force u32)val >> 7) ^ ((__force u32)val >> 14)) & mask;
		return ((__force u32)val ^
		((__force u32)val >> 7) ^
		((__force u32)val >> 14)) & mask;
		}

		static struct hlist_head *fib_hash_alloc(int bytes)
		@@ -611,7 +623,8 @@ static struct hlist_head *fib_hash_alloc(int bytes)
		return kzalloc(bytes, GFP_KERNEL);
		else
		return (struct hlist_head *)
		__get_free_pages(GFP_KERNEL \| __GFP_ZERO, get_order(bytes));
		__get_free_pages(GFP_KERNEL \| __GFP_ZERO,
		get_order(bytes));
		}

		static void fib_hash_free(struct hlist_head *hash, int bytes)
		@@ -806,7 +819,8 @@ struct fib_info fib_create_info(struct fib_config cfg)
		goto failure;
		} else {
		change_nexthops(fi) {
		if ((err = fib_check_nh(cfg, fi, nexthop_nh)) != 0)
		err = fib_check_nh(cfg, fi, nexthop_nh);
		if (err != 0)
		goto failure;
		} endfor_nexthops(fi)
		}
		@@ -819,7 +833,8 @@ struct fib_info fib_create_info(struct fib_config cfg)
		}

		link_it:
		if ((ofi = fib_find_info(fi)) != NULL) {
		ofi = fib_find_info(fi);
		if (ofi) {
		fi->fib_dead = 1;
		free_fib_info(fi);
		ofi->fib_treeref++;
		@@ -906,15 +921,14 @@ int fib_semantic_match(struct list_head head, const struct flowi flp,
		goto out_fill_res;
		}
		#else
		if (nhsel < 1) {
		if (nhsel < 1)
		goto out_fill_res;
		}
		#endif
		endfor_nexthops(fi);
		continue;

		default:
		printk(KERN_WARNING "fib_semantic_match bad type %#x\n",
		pr_warning("fib_semantic_match bad type %#x\n",
		fa->fa_type);
		return -EINVAL;
		}
		@@ -1028,10 +1042,10 @@ nla_put_failure:
		}

		/*
		Update FIB if:
		- local address disappeared -> we must delete all the entries
		referring to it.
		- device went down -> we must shutdown all nexthops going via it.
		* Update FIB if:
		* - local address disappeared -> we must delete all the entries
		* referring to it.
		* - device went down -> we must shutdown all nexthops going via it.
		*/
		int fib_sync_down_addr(struct net *net, __be32 local)
		{
		@@ -1110,10 +1124,9 @@ int fib_sync_down_dev(struct net_device *dev, int force)
		#ifdef CONFIG_IP_ROUTE_MULTIPATH

		/*
		Dead device goes up. We wake up dead nexthops.
		It takes sense only on multipath routes.
		* Dead device goes up. We wake up dead nexthops.
		* It takes sense only on multipath routes.
		*/

		int fib_sync_up(struct net_device *dev)
		{
		struct fib_info *prev_fi;
		@@ -1169,10 +1182,9 @@ int fib_sync_up(struct net_device *dev)
		}

		/*
		The algorithm is suboptimal, but it provides really
		fair weighted route distribution.
		* The algorithm is suboptimal, but it provides really
		* fair weighted route distribution.
		*/

		void fib_select_multipath(const struct flowi flp, struct fib_result res)
		{
		struct fib_info *fi = res->fi;
		@@ -1198,7 +1210,7 @@ void fib_select_multipath(const struct flowi flp, struct fib_result res)


		/* w should be random number [0..fi->fib_power-1],
		it is pretty bad approximation.
		* it is pretty bad approximation.
		*/

		w = jiffies % fi->fib_power;
		@@ -1206,7 +1218,8 @@ void fib_select_multipath(const struct flowi flp, struct fib_result res)
		change_nexthops(fi) {
		if (!(nexthop_nh->nh_flags & RTNH_F_DEAD) &&
		nexthop_nh->nh_power) {
		if ((w -= nexthop_nh->nh_power) <= 0) {
		w -= nexthop_nh->nh_power;
		if (w <= 0) {
		nexthop_nh->nh_power--;
		fi->fib_power--;
		res->nh_sel = nhsel;