[PATCH v2 net-next] bonding: don't call alb_set_slave_mac_addr() while atomic

[PATCH v2 net-next] bonding: don't call alb_set_slave_mac_addr() while atomic

[Veaceslav Falico]

alb_set_slave_mac_addr() sets the mac address in alb mode via
dev_set_mac_address(), which might sleep. It's called from
alb_handle_addr_collision_on_attach() in atomic context (under
read_lock(bond->lock)), thus triggering a bug.

Fix this by moving the lock inside alb_handle_addr_collision_on_attach().

v1->v2:
As Nikolay Aleksandrov noticed, we can drop the bond->lock completely.
Also, use bond_slave_has_mac(), when possible.

Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
---
 drivers/net/bonding/bond_alb.c | 40 +++++-----------------------------------
 1 file changed, 5 insertions(+), 35 deletions(-)

diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index a236234..27fe329 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -1175,16 +1175,13 @@ static void alb_change_hw_addr_on_detach(struct bonding *bond, struct slave *sla
  * @slave.
  *
  * assumption: this function is called before @slave is attached to the
- * 	       bond slave list.
- *
- * caller must hold the bond lock for write since the mac addresses are compared
- * and may be swapped.
+ *	       bond slave list.
  */
 static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slave *slave)
 {
-	struct slave *tmp_slave1, *tmp_slave2, *free_mac_slave;
+	struct slave *tmp_slave1, *free_mac_slave = NULL;
 	struct slave *has_bond_addr = bond->curr_active_slave;
-	int i, j, found = 0;
+	int i;
 
 	if (bond->slave_cnt == 0) {
 		/* this is the first slave */
@@ -1196,15 +1193,7 @@ static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slav
 	 * slaves in the bond.
 	 */
 	if (!ether_addr_equal_64bits(slave->perm_hwaddr, bond->dev->dev_addr)) {
-		bond_for_each_slave(bond, tmp_slave1, i) {
-			if (ether_addr_equal_64bits(tmp_slave1->dev->dev_addr,
-						    slave->dev->dev_addr)) {
-				found = 1;
-				break;
-			}
-		}
-
-		if (!found)
+		if (!bond_slave_has_mac(bond, slave->dev->dev_addr))
 			return 0;
 
 		/* Try setting slave mac to bond address and fall-through
@@ -1215,19 +1204,8 @@ static int alb_handle_addr_collision_on_attach(struct bonding *bond, struct slav
 	/* The slave's address is equal to the address of the bond.
 	 * Search for a spare address in the bond for this slave.
 	 */
-	free_mac_slave = NULL;
-
 	bond_for_each_slave(bond, tmp_slave1, i) {
-		found = 0;
-		bond_for_each_slave(bond, tmp_slave2, j) {
-			if (ether_addr_equal_64bits(tmp_slave1->perm_hwaddr,
-						    tmp_slave2->dev->dev_addr)) {
-				found = 1;
-				break;
-			}
-		}
-
-		if (!found) {
+		if (!bond_slave_has_mac(bond, tmp_slave1->perm_hwaddr)) {
 			/* no slave has tmp_slave1's perm addr
 			 * as its curr addr
 			 */
@@ -1607,15 +1585,7 @@ int bond_alb_init_slave(struct bonding *bond, struct slave *slave)
 		return res;
 	}
 
-	/* caller must hold the bond lock for write since the mac addresses
-	 * are compared and may be swapped.
-	 */
-	read_lock(&bond->lock);
-
 	res = alb_handle_addr_collision_on_attach(bond, slave);
-
-	read_unlock(&bond->lock);
-
 	if (res) {
 		return res;
 	}
-- 
1.8.1.4

Re: [PATCH v2 net-next] bonding: don't call alb_set_slave_mac_addr() while atomic

[Nikolay Aleksandrov]

On 17/06/13 19:30, Veaceslav Falico wrote:
> alb_set_slave_mac_addr() sets the mac address in alb mode via
> dev_set_mac_address(), which might sleep. It's called from
> alb_handle_addr_collision_on_attach() in atomic context (under
> read_lock(bond->lock)), thus triggering a bug.
> 
> Fix this by moving the lock inside alb_handle_addr_collision_on_attach().
> 
> v1->v2:
> As Nikolay Aleksandrov noticed, we can drop the bond->lock completely.
> Also, use bond_slave_has_mac(), when possible.
> 
> Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
> ---
>  drivers/net/bonding/bond_alb.c | 40 +++++-----------------------------------
>  1 file changed, 5 insertions(+), 35 deletions(-)
> 

Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>

Re: [PATCH v2 net-next] bonding: don't call alb_set_slave_mac_addr() while atomic

[David Miller]

From: Nikolay Aleksandrov <nikolay@redhat.com>
Date: Mon, 17 Jun 2013 19:46:39 +0200

> On 17/06/13 19:30, Veaceslav Falico wrote:
>> alb_set_slave_mac_addr() sets the mac address in alb mode via
>> dev_set_mac_address(), which might sleep. It's called from
>> alb_handle_addr_collision_on_attach() in atomic context (under
>> read_lock(bond->lock)), thus triggering a bug.
>> 
>> Fix this by moving the lock inside alb_handle_addr_collision_on_attach().
>> 
>> v1->v2:
>> As Nikolay Aleksandrov noticed, we can drop the bond->lock completely.
>> Also, use bond_slave_has_mac(), when possible.
>> 
>> Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
>> ---
>>  drivers/net/bonding/bond_alb.c | 40 +++++-----------------------------------
>>  1 file changed, 5 insertions(+), 35 deletions(-)
>> 
> 
> Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com>

Applied.