addrconf: refcnt with IPV6_PRIVACY enabled

From: Sergey Senozhatsky
Date: Tue Nov 23 2010 - 08:39:45 EST


Hello,

I've sent a patch http://lkml.org/lkml/2010/11/21/124
"[PATCH] ipv6: fix inet6_dev refcnt with IPV6_PRIVACY enabled" related
to wrong in6_dev refcnt value when IPV6_PRIVACY enabled.

I've took a second look, and question arised -
Is it actually necessary to in6_dev_hold/in6_dev_put in ipv6_regen_rndid?
In ipv6_regen_rndid we call in6_dev_hold only when mod_timer == 0,
and always in6_dev_put.


I've removed check for < 0 and goto from __ipv6_regen_rndid call, since it
always return 0.


Please, kindly review V2.

---

net/ipv6/addrconf.c | 10 +++-------
1 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 2fc35b3..8187d14 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -425,7 +425,6 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
dev->name);
ndev->cnf.use_tempaddr = -1;
} else {
- in6_dev_hold(ndev);
ipv6_regen_rndid((unsigned long) ndev);
}
#endif
@@ -1653,8 +1652,7 @@ static void ipv6_regen_rndid(unsigned long data)
if (idev->dead)
goto out;

- if (__ipv6_regen_rndid(idev) < 0)
- goto out;
+ __ipv6_regen_rndid(idev);

expires = jiffies +
idev->cnf.temp_prefered_lft * HZ -
@@ -1667,13 +1665,11 @@ static void ipv6_regen_rndid(unsigned long data)
goto out;
}

- if (!mod_timer(&idev->regen_timer, expires))
- in6_dev_hold(idev);
-
+ mod_timer(&idev->regen_timer, expires);
+
out:
write_unlock_bh(&idev->lock);
rcu_read_unlock_bh();
- in6_dev_put(idev);
}

static int __ipv6_try_regen_rndid(struct inet6_dev *idev, struct in6_addr *tmpaddr) {

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/