[PATCH iwl-net v1] i40e: Fix macvlan leak by synchronizing access to mac_filter_hash

This patch addresses a macvlan leak issue in the i40e driver caused by
concurrent access to vsi->mac_filter_hash. The leak occurs when multiple
threads attempt to modify the mac_filter_hash simultaneously, leading to
inconsistent state and potential memory leaks.

To fix this, we now wrap the calls to i40e_del_mac_filter() and zeroing
vf->default_lan_addr.addr with spin_lock/unlock_bh(&vsi->mac_filter_hash_lock),
ensuring atomic operations and preventing concurrent access.

Additionally, we add lockdep_assert_held(&vsi->mac_filter_hash_lock) in
i40e_add_mac_filter() to help catch similar issues in the future.

Reproduction steps:
1. Spawn VFs and configure port vlan on them.
2. Trigger concurrent macvlan operations (e.g., adding and deleting
	portvlan and/or mac filters).
3. Observe the potential memory leak and inconsistent state in the
	mac_filter_hash.

This synchronization ensures the integrity of the mac_filter_hash and prevents
the described leak.

Fixes: fed0d9f13266 ("i40e: Fix VF's MAC Address change on VM")
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Aleksandr Loktionov <redacted>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c        | 1 +
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 03205eb..25295ae 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c

@@ -1734,6 +1734,7 @@ struct i40e_mac_filter *i40e_add_mac_filter(struct i40e_vsi *vsi,
 	struct hlist_node *h;
 	int bkt;
 
+	lockdep_assert_held(&vsi->mac_filter_hash_lock);
 	if (vsi->info.pvid)
 		return i40e_add_filter(vsi, macaddr,
 				       le16_to_cpu(vsi->info.pvid));

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 662622f..dfa785e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c

@@ -2213,8 +2213,10 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg)
 		vfres->vsi_res[0].qset_handle
 					  = le16_to_cpu(vsi->info.qs_handle[0]);
 		if (!(vf->driver_caps & VIRTCHNL_VF_OFFLOAD_USO) && !vf->pf_set_mac) {
+			spin_lock_bh(&vsi->mac_filter_hash_lock);
 			i40e_del_mac_filter(vsi, vf->default_lan_addr.addr);
 			eth_zero_addr(vf->default_lan_addr.addr);
+			spin_unlock_bh(&vsi->mac_filter_hash_lock);
 		}
 		ether_addr_copy(vfres->vsi_res[0].default_mac_addr,
 				vf->default_lan_addr.addr);

-- 
2.25.1

On Mon, Sep 23, 2024 at 11:12:19AM +0200, Aleksandr Loktionov wrote:

This patch addresses a macvlan leak issue in the i40e driver caused by
concurrent access to vsi->mac_filter_hash. The leak occurs when multiple
threads attempt to modify the mac_filter_hash simultaneously, leading to
inconsistent state and potential memory leaks.

To fix this, we now wrap the calls to i40e_del_mac_filter() and zeroing
vf->default_lan_addr.addr with spin_lock/unlock_bh(&vsi->mac_filter_hash_lock),
ensuring atomic operations and preventing concurrent access.

Additionally, we add lockdep_assert_held(&vsi->mac_filter_hash_lock) in
i40e_add_mac_filter() to help catch similar issues in the future.

Reproduction steps:
1. Spawn VFs and configure port vlan on them.
2. Trigger concurrent macvlan operations (e.g., adding and deleting
	portvlan and/or mac filters).
3. Observe the potential memory leak and inconsistent state in the
	mac_filter_hash.

This synchronization ensures the integrity of the mac_filter_hash and prevents
the described leak.

Fixes: fed0d9f13266 ("i40e: Fix VF's MAC Address change on VM")
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Aleksandr Loktionov <redacted>

Thanks Aleksandr,

I see that:

1) All calls to i40e_add_mac_filter() and all other calls
   to i40e_del_mac_filter() are already protected by
   vsi->mac_filter_hash_lock.

2) i40e_del_mac_filter() already asserts that
   vsi->mac_filter_hash_lock is held.

So this looks good to me.

Reviewed-by: Simon Horman <horms@kernel.org>

...

-----Original Message-----
From: Intel-wired-lan <redacted> On Behalf Of Simon
Horman
Sent: Tuesday, September 24, 2024 8:57 AM
To: Loktionov, Aleksandr <redacted>
Cc: Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Kubalewski, Arkadiusz
[off-list ref]; intel-wired-lan@lists.osuosl.org;
netdev@vger.kernel.org
Subject: Re: [Intel-wired-lan] [PATCH iwl-net v1] i40e: Fix macvlan leak by
synchronizing access to mac_filter_hash

On Mon, Sep 23, 2024 at 11:12:19AM +0200, Aleksandr Loktionov wrote:

quoted

This patch addresses a macvlan leak issue in the i40e driver caused by
concurrent access to vsi->mac_filter_hash. The leak occurs when
multiple threads attempt to modify the mac_filter_hash simultaneously,
leading to inconsistent state and potential memory leaks.

To fix this, we now wrap the calls to i40e_del_mac_filter() and
zeroing
vf->default_lan_addr.addr with
vf->spin_lock/unlock_bh(&vsi->mac_filter_hash_lock),
ensuring atomic operations and preventing concurrent access.

Additionally, we add lockdep_assert_held(&vsi->mac_filter_hash_lock)
in
i40e_add_mac_filter() to help catch similar issues in the future.

Reproduction steps:
1. Spawn VFs and configure port vlan on them.
2. Trigger concurrent macvlan operations (e.g., adding and deleting
	portvlan and/or mac filters).
3. Observe the potential memory leak and inconsistent state in the
	mac_filter_hash.

This synchronization ensures the integrity of the mac_filter_hash and
prevents the described leak.

Fixes: fed0d9f13266 ("i40e: Fix VF's MAC Address change on VM")
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Signed-off-by: Aleksandr Loktionov <redacted>

Thanks Aleksandr,

I see that:

1) All calls to i40e_add_mac_filter() and all other calls
   to i40e_del_mac_filter() are already protected by
   vsi->mac_filter_hash_lock.

2) i40e_del_mac_filter() already asserts that
   vsi->mac_filter_hash_lock is held.

So this looks good to me.

Reviewed-by: Simon Horman <horms@kernel.org>

...

Tested-by: Rafal Romanowski <redacted>