ip rule duplicates

Hi,

When we add the same rule again with flag NLM_F_EXCL we expect that we receive error:
RTNETLINK answers: File exists
This behaviour is already in ip routing part.

I have noticed that iproute2 when adds new rule it attach flag NLM_F_EXCL to call. (see http://git.kernel.org/cgit/linux/kernel/git/shemminger/iproute2.git/tree/ip/iprule.c#n334)

Next thing what I found is that this flag is not handled from kernel side.

I implemented this feature and I tested this with qemu x86 on:
linux-4.5.4
linux (git)
linux-stable (git)
Tested with ipv4 and ipv6.

current behaviour with ipv4:
localhost ~ # ip rule
0: from all lookup local
32766: from all lookup main
32767: from all lookup default
localhost ~ # ip rule add from 10.46.177.97 lookup 104 pref 1005 localhost ~ # ip rule add from 10.46.177.97 lookup 104 pref 1005 localhost ~ # ip rule
0: from all lookup local
1005: from 10.46.177.97 lookup 104
1005: from 10.46.177.97 lookup 104
32766: from all lookup main
32767: from all lookup default

expected behavior after patch:
localhost ~ # ip rule
0: from all lookup local
32766: from all lookup main
32767: from all lookup default
localhost ~ # ip rule add from 10.46.177.97 lookup 104 pref 1005 localhost ~ # ip rule add from 10.46.177.97 lookup 104 pref 1005 RTNETLINK answers: File exists
localhost ~ # ip rule
0: from all lookup local
1005: from 10.46.177.97 lookup 104
32766: from all lookup main
32767: from all lookup default

There was already topic regarding this but I don't see any changes merged and problem still occurs. (see http://marc.info/?l=linux-netdev&m=113577886110391&w=2)

Best regards,

Mateusz Bajorski

From: Mateusz Bajorski <[email protected]>
Date: Fri, 20 May 2016 14:29:56 +0200
Subject: [PATCH] Added NLM_F_EXCL support to fib_nl_newrule

---
 net/core/fib_rules.c | 41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

index 840aceb..c1bc07cd 100644
@@ -291,6 +291,47 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh)
 	if (err < 0)
 		goto errout;
 
+	if (nlh->nlmsg_flags & NLM_F_EXCL) {
+		list_for_each_entry(rule, &ops->rules_list, list) {
+			if (frh->action && (frh->action != rule->action))
+				continue;
+
+			if (frh_get_table(frh, tb) &&
+				(frh_get_table(frh, tb) != rule->table))
+				continue;
+
+			if (tb[FRA_PRIORITY] &&
+				(rule->pref != nla_get_u32(tb[FRA_PRIORITY])))
+				continue;
+
+			if (tb[FRA_IIFNAME] &&
+				nla_strcmp(tb[FRA_IIFNAME], rule->iifname))
+				continue;
+
+			if (tb[FRA_OIFNAME] &&
+				nla_strcmp(tb[FRA_OIFNAME], rule->oifname))
+				continue;
+
+			if (tb[FRA_FWMARK] &&
+				(rule->mark != nla_get_u32(tb[FRA_FWMARK])))
+				continue;
+
+			if (tb[FRA_FWMASK] &&
+				(rule->mark_mask != nla_get_u32(tb[FRA_FWMASK])))
+				continue;
+
+			if (tb[FRA_TUN_ID] &&
+				(rule->tun_id != nla_get_be64(tb[FRA_TUN_ID])))
+				continue;
+
+			if (!ops->compare(rule, frh, tb))
+				continue;
+
+			err = -EEXIST;
+			goto errout;
+		}
+	}
+
 	rule = kzalloc(ops->rule_size, GFP_KERNEL);
 	if (rule == NULL) {
 		err = -ENOMEM;

2.6.4