netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] net_sched: return correct value for *notify* functions
@ 2019-03-11 10:15 Zhike Wang
  2019-03-13  5:48 ` Cong Wang
  2019-03-13 20:48 ` David Miller
  0 siblings, 2 replies; 3+ messages in thread
From: Zhike Wang @ 2019-03-11 10:15 UTC (permalink / raw)
  To: jhs, xiyou.wangcong, jiri, netdev; +Cc: Zhike Wang

It is confusing to directly use return value of netlink_send()/
netlink_unicast() as the return value of *notify*, as it may be not
error at all.

Example: in tc_del_tfilter(), after calling tfilter_del_notify(), it will
goto errout if (err). However, the netlink_send()/netlink_unicast() will
return positive value even for successful case. So it may not call
tcf_chain_tp_remove() and so on to clean up the resource, as a result,
resource is leaked.

It may be easier to only check the return value of tfilter_del_nofiy(),
but it is more clean to correct all related functions.

Co-developed-by: Zengmo Gao <gaozengmo@jd.com>
Signed-off-by: Zhike Wang <wangzhike@jd.com>
---
 net/sched/cls_api.c | 32 +++++++++++++++++++++++---------
 net/sched/sch_api.c | 15 +++++++++++----
 2 files changed, 34 insertions(+), 13 deletions(-)

diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 478095d..15cac37 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -1896,6 +1896,7 @@ static int tfilter_notify(struct net *net, struct sk_buff *oskb,
 {
 	struct sk_buff *skb;
 	u32 portid = oskb ? NETLINK_CB(oskb).portid : 0;
+	int err = 0;
 
 	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (!skb)
@@ -1909,10 +1910,14 @@ static int tfilter_notify(struct net *net, struct sk_buff *oskb,
 	}
 
 	if (unicast)
-		return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT);
+		err = netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT);
+	else
+		err = rtnetlink_send(skb, net, portid, RTNLGRP_TC,
+				     n->nlmsg_flags & NLM_F_ECHO);
 
-	return rtnetlink_send(skb, net, portid, RTNLGRP_TC,
-			      n->nlmsg_flags & NLM_F_ECHO);
+	if (err > 0)
+		err = 0;
+	return err;
 }
 
 static int tfilter_del_notify(struct net *net, struct sk_buff *oskb,
@@ -1944,12 +1949,15 @@ static int tfilter_del_notify(struct net *net, struct sk_buff *oskb,
 	}
 
 	if (unicast)
-		return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT);
-
-	err = rtnetlink_send(skb, net, portid, RTNLGRP_TC,
-			     n->nlmsg_flags & NLM_F_ECHO);
+		err = netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT);
+	else
+		err = rtnetlink_send(skb, net, portid, RTNLGRP_TC,
+				     n->nlmsg_flags & NLM_F_ECHO);
 	if (err < 0)
 		NL_SET_ERR_MSG(extack, "Failed to send filter delete notification");
+
+	if (err > 0)
+		err = 0;
 	return err;
 }
 
@@ -2691,6 +2699,7 @@ static int tc_chain_notify(struct tcf_chain *chain, struct sk_buff *oskb,
 	struct tcf_block *block = chain->block;
 	struct net *net = block->net;
 	struct sk_buff *skb;
+	int err = 0;
 
 	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (!skb)
@@ -2704,9 +2713,14 @@ static int tc_chain_notify(struct tcf_chain *chain, struct sk_buff *oskb,
 	}
 
 	if (unicast)
-		return netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT);
+		err = netlink_unicast(net->rtnl, skb, portid, MSG_DONTWAIT);
+	else
+		err = rtnetlink_send(skb, net, portid, RTNLGRP_TC,
+				     flags & NLM_F_ECHO);
 
-	return rtnetlink_send(skb, net, portid, RTNLGRP_TC, flags & NLM_F_ECHO);
+	if (err > 0)
+		err = 0;
+	return err;
 }
 
 static int tc_chain_notify_delete(const struct tcf_proto_ops *tmplt_ops,
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 352b46f..fb8f138 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -1824,6 +1824,7 @@ static int tclass_notify(struct net *net, struct sk_buff *oskb,
 {
 	struct sk_buff *skb;
 	u32 portid = oskb ? NETLINK_CB(oskb).portid : 0;
+	int err = 0;
 
 	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (!skb)
@@ -1834,8 +1835,11 @@ static int tclass_notify(struct net *net, struct sk_buff *oskb,
 		return -EINVAL;
 	}
 
-	return rtnetlink_send(skb, net, portid, RTNLGRP_TC,
-			      n->nlmsg_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;
 }
 
 static int tclass_del_notify(struct net *net,
@@ -1866,8 +1870,11 @@ static int tclass_del_notify(struct net *net,
 		return err;
 	}
 
-	return rtnetlink_send(skb, net, portid, RTNLGRP_TC,
-			      n->nlmsg_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;
 }
 
 #ifdef CONFIG_NET_CLS
-- 
1.8.3.1


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] net_sched: return correct value for *notify* functions
  2019-03-11 10:15 [PATCH] net_sched: return correct value for *notify* functions Zhike Wang
@ 2019-03-13  5:48 ` Cong Wang
  2019-03-13 20:48 ` David Miller
  1 sibling, 0 replies; 3+ messages in thread
From: Cong Wang @ 2019-03-13  5:48 UTC (permalink / raw)
  To: Zhike Wang; +Cc: Jamal Hadi Salim, Jiri Pirko, Linux Kernel Network Developers

On Mon, Mar 11, 2019 at 3:16 AM Zhike Wang <wangzhike@jd.com> wrote:
>
> It is confusing to directly use return value of netlink_send()/
> netlink_unicast() as the return value of *notify*, as it may be not
> error at all.
>
> Example: in tc_del_tfilter(), after calling tfilter_del_notify(), it will
> goto errout if (err). However, the netlink_send()/netlink_unicast() will
> return positive value even for successful case. So it may not call
> tcf_chain_tp_remove() and so on to clean up the resource, as a result,
> resource is leaked.
>
> It may be easier to only check the return value of tfilter_del_nofiy(),
> but it is more clean to correct all related functions.


Looks good to me, as tc action already uses it in a same way.

Acked-by: Cong Wang <xiyou.wangcong@gmail.com>

Thanks.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] net_sched: return correct value for *notify* functions
  2019-03-11 10:15 [PATCH] net_sched: return correct value for *notify* functions Zhike Wang
  2019-03-13  5:48 ` Cong Wang
@ 2019-03-13 20:48 ` David Miller
  1 sibling, 0 replies; 3+ messages in thread
From: David Miller @ 2019-03-13 20:48 UTC (permalink / raw)
  To: wangzhike; +Cc: jhs, xiyou.wangcong, jiri, netdev

From: Zhike Wang <wangzhike@jd.com>
Date: Mon, 11 Mar 2019 03:15:54 -0700

> It is confusing to directly use return value of netlink_send()/
> netlink_unicast() as the return value of *notify*, as it may be not
> error at all.
> 
> Example: in tc_del_tfilter(), after calling tfilter_del_notify(), it will
> goto errout if (err). However, the netlink_send()/netlink_unicast() will
> return positive value even for successful case. So it may not call
> tcf_chain_tp_remove() and so on to clean up the resource, as a result,
> resource is leaked.
> 
> It may be easier to only check the return value of tfilter_del_nofiy(),
> but it is more clean to correct all related functions.
> 
> Co-developed-by: Zengmo Gao <gaozengmo@jd.com>
> Signed-off-by: Zhike Wang <wangzhike@jd.com>

Applied.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2019-03-13 20:48 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-03-11 10:15 [PATCH] net_sched: return correct value for *notify* functions Zhike Wang
2019-03-13  5:48 ` Cong Wang
2019-03-13 20:48 ` David Miller

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).