i40e: fix race condition by adding filter's intermediate sync state

[ Upstream commit f30490e9695ef7da3d0899c6a0293cc7cd373567 ]

Fix a race condition in the i40e driver that leads to MAC/VLAN filters
becoming corrupted and leaking. Address the issue that occurs under
heavy load when multiple threads are concurrently modifying MAC/VLAN
filters by setting mac and port VLAN.

1. Thread T0 allocates a filter in i40e_add_filter() within
        i40e_ndo_set_vf_port_vlan().
2. Thread T1 concurrently frees the filter in __i40e_del_filter() within
        i40e_ndo_set_vf_mac().
3. Subsequently, i40e_service_task() calls i40e_sync_vsi_filters(), which
        refers to the already freed filter memory, causing corruption.

Reproduction steps:
1. Spawn multiple VFs.
2. Apply a concurrent heavy load by running parallel operations to change
        MAC addresses on the VFs and change port VLANs on the host.
3. Observe errors in dmesg:
"Error I40E_AQ_RC_ENOSPC adding RX filters on VF XX,
	please set promiscuous on manually for VF XX".

Exact code for stable reproduction Intel can't open-source now.

The fix involves implementing a new intermediate filter state,
I40E_FILTER_NEW_SYNC, for the time when a filter is on a tmp_add_list.
These filters cannot be deleted from the hash list directly but
must be removed using the full process.

Fixes: 278e7d0b9d68 ("i40e: store MAC/VLAN filters in a hash with the MAC Address as key")
Signed-off-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Reviewed-by: Michal Schmidt <mschmidt@redhat.com>
Tested-by: Michal Schmidt <mschmidt@redhat.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

3 files changed, 12 insertions, 2 deletions

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 3e6839ac1f0f..68f403dd2f52 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -763,6 +763,7 @@ enum i40e_filter_state {
 	I40E_FILTER_ACTIVE,		/* Added to switch by FW */
 	I40E_FILTER_FAILED,		/* Rejected by FW */
 	I40E_FILTER_REMOVE,		/* To be removed */
+	I40E_FILTER_NEW_SYNC,		/* New, not sent yet, is in i40e_sync_vsi_filters() */
 /* There is no 'removed' state; the filter struct is freed */
 };
 struct i40e_mac_filter {
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 999c9708def5..a2fca58a91c3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -106,6 +106,7 @@ static char *i40e_filter_state_string[] = {
 	"ACTIVE",
 	"FAILED",
 	"REMOVE",
+	"NEW_SYNC",
 };
 
 /**
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 80472aa1deba..806cec458a07 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1265,6 +1265,7 @@ int i40e_count_filters(struct i40e_vsi *vsi)
 
 	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
 		if (f->state == I40E_FILTER_NEW ||
+		    f->state == I40E_FILTER_NEW_SYNC ||
 		    f->state == I40E_FILTER_ACTIVE)
 			++cnt;
 	}
@@ -1451,6 +1452,8 @@ static int i40e_correct_mac_vlan_filters(struct i40e_vsi *vsi,
 
 			new->f = add_head;
 			new->state = add_head->state;
+			if (add_head->state == I40E_FILTER_NEW)
+				add_head->state = I40E_FILTER_NEW_SYNC;
 
 			/* Add the new filter to the tmp list */
 			hlist_add_head(&new->hlist, tmp_add_list);
@@ -1560,6 +1563,8 @@ static int i40e_correct_vf_mac_vlan_filters(struct i40e_vsi *vsi,
 				return -ENOMEM;
 			new_mac->f = add_head;
 			new_mac->state = add_head->state;
+			if (add_head->state == I40E_FILTER_NEW)
+				add_head->state = I40E_FILTER_NEW_SYNC;
 
 			/* Add the new filter to the tmp list */
 			hlist_add_head(&new_mac->hlist, tmp_add_list);
@@ -2445,7 +2450,8 @@ static int
 i40e_aqc_broadcast_filter(struct i40e_vsi *vsi, const char *vsi_name,
 			  struct i40e_mac_filter *f)
 {
-	bool enable = f->state == I40E_FILTER_NEW;
+	bool enable = f->state == I40E_FILTER_NEW ||
+		      f->state == I40E_FILTER_NEW_SYNC;
 	struct i40e_hw *hw = &vsi->back->hw;
 	int aq_ret;
 
@@ -2619,6 +2625,7 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 
 				/* Add it to the hash list */
 				hlist_add_head(&new->hlist, &tmp_add_list);
+				f->state = I40E_FILTER_NEW_SYNC;
 			}
 
 			/* Count the number of active (current and new) VLAN
@@ -2770,7 +2777,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 		spin_lock_bh(&vsi->mac_filter_hash_lock);
 		hlist_for_each_entry_safe(new, h, &tmp_add_list, hlist) {
 			/* Only update the state if we're still NEW */
-			if (new->f->state == I40E_FILTER_NEW)
+			if (new->f->state == I40E_FILTER_NEW ||
+			    new->f->state == I40E_FILTER_NEW_SYNC)
 				new->f->state = new->state;
 			hlist_del(&new->hlist);
 			netdev_hw_addr_refcnt(new->f, vsi->netdev, -1);