Netdev List
 help / color / mirror / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Gao feng <gaofeng@cn.fujitsu.com>
Cc: Neil Horman <nhorman@tuxdriver.com>,
	netdev@vger.kernel.org, David Miller <davem@davemloft.net>,
	Jiri Bohac <jbohac@suse.cz>
Subject: Re: [RFC PATCH] ipv6: Split from and expires field in dst_entry out of union [net-next]
Date: Tue, 19 Feb 2013 23:00:10 -0800	[thread overview]
Message-ID: <1361343610.19353.177.camel@edumazet-glaptop> (raw)
In-Reply-To: <51246E43.5010202@cn.fujitsu.com>

On Wed, 2013-02-20 at 14:33 +0800, Gao feng wrote:

> How can we?
> one usage of rt6_update_expires and rt6_set_expires
> is changing rt6->dst.from to rt6->dst.expires, we should release the
> already holded reference of rt6->dst.from.
> 

Just don't union the two fields, as Neil did.

Setting the dst.expires value should not change dst.from at all.

Something like the (untested, because its too late here) patch ?

 include/net/dst.h     |   11 +++++------
 include/net/ip6_fib.h |   21 +++------------------
 net/ipv6/route.c      |    3 +--
 3 files changed, 9 insertions(+), 26 deletions(-)

diff --git a/include/net/dst.h b/include/net/dst.h
index 3da47e0..3f31a48 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -36,13 +36,12 @@ struct dst_entry {
 	struct net_device       *dev;
 	struct  dst_ops	        *ops;
 	unsigned long		_metrics;
-	union {
-		unsigned long           expires;
-		/* point to where the dst_entry copied from */
-		struct dst_entry        *from;
-	};
+	unsigned long           expires;
+
+	/* point to where the dst_entry copied from */
+	struct dst_entry        *from;
+
 	struct dst_entry	*path;
-	void			*__pad0;
 #ifdef CONFIG_XFRM
 	struct xfrm_state	*xfrm;
 #else
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 6919a50..731b35c 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -164,33 +164,19 @@ static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst)
 
 static inline void rt6_clean_expires(struct rt6_info *rt)
 {
-	if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from)
-		dst_release(rt->dst.from);
-
 	rt->rt6i_flags &= ~RTF_EXPIRES;
-	rt->dst.from = NULL;
+	rt->dst.expires = 0;
 }
 
 static inline void rt6_set_expires(struct rt6_info *rt, unsigned long expires)
 {
-	if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from)
-		dst_release(rt->dst.from);
-
 	rt->rt6i_flags |= RTF_EXPIRES;
 	rt->dst.expires = expires;
 }
 
 static inline void rt6_update_expires(struct rt6_info *rt, int timeout)
 {
-	if (!(rt->rt6i_flags & RTF_EXPIRES)) {
-		if (rt->dst.from)
-			dst_release(rt->dst.from);
-		/* dst_set_expires relies on expires == 0 
-		 * if it has not been set previously.
-		 */
-		rt->dst.expires = 0;
-	}
-
+	rt->dst.expires = 0;
 	dst_set_expires(&rt->dst, timeout);
 	rt->rt6i_flags |= RTF_EXPIRES;
 }
@@ -199,13 +185,12 @@ static inline void rt6_set_from(struct rt6_info *rt, struct rt6_info *from)
 {
 	struct dst_entry *new = (struct dst_entry *) from;
 
-	if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from) {
+	if (rt->dst.from) {
 		if (new == rt->dst.from)
 			return;
 		dst_release(rt->dst.from);
 	}
 
-	rt->rt6i_flags &= ~RTF_EXPIRES;
 	rt->dst.from = new;
 	dst_hold(new);
 }
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 515bb51..6874b6e 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -296,8 +296,7 @@ static void ip6_dst_destroy(struct dst_entry *dst)
 		in6_dev_put(idev);
 	}
 
-	if (!(rt->rt6i_flags & RTF_EXPIRES) && dst->from)
-		dst_release(dst->from);
+	dst_release(dst->from);
 
 	if (rt6_has_peer(rt)) {
 		struct inet_peer *peer = rt6_peer_ptr(rt);

  reply	other threads:[~2013-02-20  7:00 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-18 23:55 [RFC] ipv6: rt6_update_expires() seems racy Eric Dumazet
2013-02-19 20:28 ` [RFC PATCH] ipv6: Split from and expires field in dst_entry out of union [net-next] Neil Horman
2013-02-19 21:17   ` Eric Dumazet
2013-02-19 21:49     ` Neil Horman
2013-02-19 21:55       ` Eric Dumazet
2013-02-20  6:33         ` Gao feng
2013-02-20  7:00           ` Eric Dumazet [this message]
2013-02-20  9:31             ` Gao feng
2013-02-20 10:02             ` YOSHIFUJI Hideaki
2013-02-20 10:17               ` [PATCH] ipv6: fix race condition regarding dst->expires and dst->from YOSHIFUJI Hideaki
2013-02-20 10:25                 ` YOSHIFUJI Hideaki
2013-02-20 10:29                   ` [PATCH V2] " YOSHIFUJI Hideaki
2013-02-20 16:12                     ` Eric Dumazet
2013-02-20 16:34                       ` Neil Horman
2013-02-20 20:12                     ` David Miller
2013-02-20 10:55   ` [RFC PATCH] ipv6: Split from and expires field in dst_entry out of union [net-next] David Laight
2013-02-20 12:02     ` Neil Horman
2013-02-20 12:08       ` David Laight

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1361343610.19353.177.camel@edumazet-glaptop \
    --to=eric.dumazet@gmail.com \
    --cc=davem@davemloft.net \
    --cc=gaofeng@cn.fujitsu.com \
    --cc=jbohac@suse.cz \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox