summaryrefslogtreecommitdiff
path: root/net/ipv6/ip6_flowlabel.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/ipv6/ip6_flowlabel.c')
-rw-r--r--net/ipv6/ip6_flowlabel.c39
1 files changed, 23 insertions, 16 deletions
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index 78d1d913e36c..eb7a940310f4 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -611,6 +611,7 @@ done:
#ifdef CONFIG_PROC_FS
struct ip6fl_iter_state {
+ struct seq_net_private p;
int bucket;
};
@@ -620,12 +621,15 @@ static struct ip6_flowlabel *ip6fl_get_first(struct seq_file *seq)
{
struct ip6_flowlabel *fl = NULL;
struct ip6fl_iter_state *state = ip6fl_seq_private(seq);
+ struct net *net = seq_file_net(seq);
for (state->bucket = 0; state->bucket <= FL_HASH_MASK; ++state->bucket) {
- if (fl_ht[state->bucket]) {
- fl = fl_ht[state->bucket];
+ fl = fl_ht[state->bucket];
+
+ while (fl && fl->fl_net != net)
+ fl = fl->next;
+ if (fl)
break;
- }
}
return fl;
}
@@ -633,12 +637,18 @@ static struct ip6_flowlabel *ip6fl_get_first(struct seq_file *seq)
static struct ip6_flowlabel *ip6fl_get_next(struct seq_file *seq, struct ip6_flowlabel *fl)
{
struct ip6fl_iter_state *state = ip6fl_seq_private(seq);
+ struct net *net = seq_file_net(seq);
fl = fl->next;
+try_again:
+ while (fl && fl->fl_net != net)
+ fl = fl->next;
+
while (!fl) {
- if (++state->bucket <= FL_HASH_MASK)
+ if (++state->bucket <= FL_HASH_MASK) {
fl = fl_ht[state->bucket];
- else
+ goto try_again;
+ } else
break;
}
return fl;
@@ -708,8 +718,8 @@ static const struct seq_operations ip6fl_seq_ops = {
static int ip6fl_seq_open(struct inode *inode, struct file *file)
{
- return seq_open_private(file, &ip6fl_seq_ops,
- sizeof(struct ip6fl_iter_state));
+ return seq_open_net(inode, file, &ip6fl_seq_ops,
+ sizeof(struct ip6fl_iter_state));
}
static const struct file_operations ip6fl_seq_fops = {
@@ -717,12 +727,13 @@ static const struct file_operations ip6fl_seq_fops = {
.open = ip6fl_seq_open,
.read = seq_read,
.llseek = seq_lseek,
- .release = seq_release_private,
+ .release = seq_release_net,
};
static int ip6_flowlabel_proc_init(struct net *net)
{
- if (!proc_net_fops_create(net, "ip6_flowlabel", S_IRUGO, &ip6fl_seq_fops))
+ if (!proc_net_fops_create(net, "ip6_flowlabel",
+ S_IRUGO, &ip6fl_seq_fops))
return -ENOMEM;
return 0;
}
@@ -745,25 +756,21 @@ static inline void ip6_flowlabel_proc_fini(struct net *net)
static inline void ip6_flowlabel_net_exit(struct net *net)
{
ip6_fl_purge(net);
+ ip6_flowlabel_proc_fini(net);
}
static struct pernet_operations ip6_flowlabel_net_ops = {
+ .init = ip6_flowlabel_proc_init,
.exit = ip6_flowlabel_net_exit,
};
int ip6_flowlabel_init(void)
{
- int err;
-
- err = register_pernet_subsys(&ip6_flowlabel_net_ops);
- if (err)
- return err;
- return ip6_flowlabel_proc_init(&init_net);
+ return register_pernet_subsys(&ip6_flowlabel_net_ops);
}
void ip6_flowlabel_cleanup(void)
{
del_timer(&ip6_fl_gc_timer);
unregister_pernet_subsys(&ip6_flowlabel_net_ops);
- ip6_flowlabel_proc_fini(&init_net);
}