netdev - Re: [PATCH net-next 2/3] bonding: add new slave param and bond_slave_state

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 17 Feb 2014 18:07:06 -0800
From:	Jay Vosburgh <fubar@...ibm.com>
To:	Ding Tianhong <dingtianhong@...wei.com>
cc:	vfalico@...hat.com, andy@...yhouse.net, cwang@...pensource.com,
	jiri@...nulli.us, thomas@...nzmann.de, eric.dumazet@...il.com,
	sfeldma@...ulusnetworks.com, davem@...emloft.net,
	netdev@...r.kernel.org
Subject: Re: [PATCH net-next 2/3] bonding: add new slave param and bond_slave_state_notify()

Ding Tianhong <dingtianhong@...wei.com> wrote:

>Add a new slave parameter which called should_notify, if the slave's state
>changed and don't notify yet, the parameter will be set to 1, and then if
>the slave's state changed again, the param will be set to 0, it indicate that
>the slave's state has been restored, no need to notify any one.
>
>The bond_slave_state_notify() will check whether the status changed and then
>decide to notify or not.
>
>Cc: Jay Vosburgh <fubar@...ibm.com>
>Cc: Veaceslav Falico <vfalico@...hat.com>
>Cc: Andy Gospodarek <andy@...yhouse.net>
>Signed-off-by: Ding Tianhong <dingtianhong@...wei.com>
>---
> drivers/net/bonding/bonding.h | 44 +++++++++++++++++++++++++++++++++++++++++--
> 1 file changed, 42 insertions(+), 2 deletions(-)
>
>diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
>index d210124..4d0cd41 100644
>--- a/drivers/net/bonding/bonding.h
>+++ b/drivers/net/bonding/bonding.h
>@@ -195,7 +195,8 @@ struct slave {
> 	s8     new_link;
> 	u8     backup:1,   /* indicates backup slave. Value corresponds with
> 			      BOND_STATE_ACTIVE and BOND_STATE_BACKUP */
>-	       inactive:1; /* indicates inactive slave */
>+	       inactive:1, /* indicates inactive slave */
>+	       should_notify:1; /* indicateds whether the state changed */
> 	u8     duplex;
> 	u32    original_mtu;
> 	u32    link_failure_count;
>@@ -311,8 +312,47 @@ static inline void bond_set_slave_state(struct slave *slave,
> 	else
> 		return;
>
>-	if (notify)
>+	if (notify) {
> 		rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_KERNEL);
>+		slave->should_notify = 0;
>+	} else {
>+		if (slave->should_notify)
>+			slave->should_notify = 0;
>+		else
>+			slave->should_notify = 1;
>+	}
>+}
>+
>+static inline void bond_slave_state_notify(struct bonding *bond,
>+					   bool rtnl_locked)
>+{
>+	struct list_head *iter;
>+	struct slave *tmp;
>+
>+	rcu_read_lock();
>+	bond_for_each_slave_rcu(bond, tmp, iter) {
>+		if (tmp->should_notify) {
>+			rcu_read_unlock();
>+			goto should_notify;
>+		}
>+	}
>+	rcu_read_unlock();
>+	return;
>+
>+should_notify:
>+
>+	if (!rtnl_locked && !rtnl_trylock())
>+		return;
>+
>+	bond_for_each_slave(bond, tmp, iter) {
>+		if (tmp->should_notify) {
>+			rtmsg_ifinfo(RTM_NEWLINK, tmp->dev, 0, GFP_KERNEL);
>+			tmp->should_notify = 0;
>+		}
>+	}
>+
>+	if (!rtnl_locked)
>+		rtnl_unlock();
> }

	This function (bond_slave_state_notify) seems overly complicated
given that there appears to be only one caller.  In particular, why
bother with the "rtnl_locked" flag at all, when it is never called with
it set to true?  Really, with only one caller (in patch 3 of the
series), I'm not convinced this even needs to be a separate function.

	-J

>
> static inline void bond_slave_state_change(struct bonding *bond)
>-- 
>1.8.0

---
	-Jay Vosburgh, IBM Linux Technology Center, fubar@...ibm.com

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html