/*
* ip6_flowlabel.c IPv6 flowlabel manager.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
* Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
*/
#include <linux/capability.h>
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/net.h>
#include <linux/netdevice.h>
#include <linux/in6.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/slab.h>
#include <linux/export.h>
#include <linux/pid_namespace.h>
#include <net/net_namespace.h>
#include <net/sock.h>
#include <net/ipv6.h>
#include <net/rawv6.h>
#include <net/transp_v6.h>
#include <linux/uaccess.h>
#define FL_MIN_LINGER 6 /* Minimal linger. It is set to 6sec specified
in old IPv6 RFC. Well, it was reasonable value.
*/
#define FL_MAX_LINGER 150 /* Maximal linger timeout */
/* FL hash table */
#define FL_MAX_PER_SOCK 32
#define FL_MAX_SIZE 4096
#define FL_HASH_MASK 255
#define FL_HASH(l) (ntohl(l)&FL_HASH_MASK)
static atomic_t fl_size = ATOMIC_INIT(0);
static struct ip6_flowlabel __rcu *fl_ht[FL_HASH_MASK+1];
static void ip6_fl_gc(unsigned long dummy);
static DEFINE_TIMER(ip6_fl_gc_timer, ip6_fl_gc, 0, 0);
/* FL hash table lock: it protects only of GC */
static DEFINE_SPINLOCK(ip6_fl_lock);
/* Big socket sock */
static DEFINE_SPINLOCK(ip6_sk_fl_lock);
#define for_each_fl_rcu(hash, fl) \
for (fl = rcu_dereference_bh(fl_ht[(hash)]); \
fl != NULL; \
fl = rcu_dereference_bh(fl->next))
#define for_each_fl_continue_rcu(fl) \
for (fl = rcu_dereference_bh(fl->next); \
fl != NULL; \
fl = rcu_dereference_bh(fl->next))
#define for_each_sk_fl_rcu(np, sfl) \
for (sfl = rcu_dereference_bh(np->ipv6_fl_list); \
sfl != NULL; \
sfl = rcu_dereference_bh(sfl->next))
static inline struct ip6_flowlabel *__fl_lookup(struct net *net, __be32 label)
{
struct ip6_flowlabel *fl;
for_each_fl_rcu(FL_HASH(label), fl) {
if (fl->label == label && net_eq(fl->fl_net, net))
return fl;
}
return NULL;
}
static struct ip6_flowlabel *fl_lookup(struct net *net, __be32 label)
{
struct ip6_flowlabel *fl;
rcu_read_lock_bh();
fl = __fl_lookup(net, label);
if (fl && !atomic_inc_not_zero(&fl->users))
fl = NULL;
rcu_read_unlock_bh();
return fl;
}
static void fl_free(struct ip6_flowlabel *fl)
{
if (fl) {
if (fl->share == IPV6_FL_S_PROCESS)
put_pid(fl->owner.pid);
kfree(fl->opt);
kfree_rcu(fl, rcu);
}
}
static void fl_release(struct ip6_flowlabel *fl)
{
spin_lock_bh(&ip6_fl_lock);
fl->lastuse = jiffies;
if (atomic_dec_and_test(&fl->users)) {
unsigned long ttd = fl->lastuse + fl->linger;
if (time_after(ttd, fl->expires))
fl->expires = ttd;
ttd = fl->expires;
if (fl->opt && fl->share == IPV6_FL_S_EXCL) {
struct ipv6_txoptions *opt = fl->opt;
fl->opt = NULL;
kfree(opt);
}
if (!timer_pending(&ip6_fl_gc_timer) ||
time_after(ip6_fl_gc_timer.expires, ttd))
mod_timer(&ip6_fl_gc_timer, ttd);
}
spin_unlock_bh(&ip6_fl_lock);
}
static void ip6_fl_gc(unsigned long dummy)
{
int i;
unsigned long now = jiffies;
unsigned long sched = 0;
spin_lock(&ip6_fl_lock);
for (i = 0; i <= FL_HASH_MASK; i++) {
struct ip6_flowlabel *fl;
struct ip6_flowlabel __rcu **flp;
flp = &fl_ht[i];
while ((fl = rcu_dereference_protected(*flp,
lockdep_is_held(&ip6_fl_lock))) != NULL) {
if (atomic_read(&fl->users) == 0) {
unsigned long ttd = fl->lastuse + fl->linger;
if (time_after(ttd, fl->expires))
fl->expires = ttd;
ttd = fl->expires;
if (time_after_eq(now, ttd)) {
*flp = fl->next;
fl_free(fl);
atomic_dec(&fl_size);
continue;
}
if (!sched || time_before(ttd, sched))
sched = ttd;
}
flp = &fl->next;
}
}
if (!sched && atomic_read(&fl_size))
sched = now + FL_MAX_LINGER;
if (sched) {
mod_timer(&ip6_fl_gc_timer, sched);
}
spin_unlock(&ip6_fl_lock);
}
static void __net_exit ip6_fl_purge(struct net *net)
{
int i;
spin_lock_bh(&ip6_fl_lock);
for (i = 0; i <= FL_HASH_MASK; i++) {
struct ip6_flowlabel *fl;
struct ip6_flowlabel __rcu **flp;
flp = &fl_ht[i];
while ((fl =