[PATCH net 0/7][pull request] Intel Wired LAN Driver Updates 2022-07-11

[PATCH net 0/7][pull request] Intel Wired LAN Driver Updates 2022-07-11

[Tony Nguyen]

This series contains updates to iavf driver only.

Przemyslaw fixes handling of multiple VLAN requests to account for
individual errors instead of rejecting them all. He removes incorrect
implementations of ETHTOOL_COALESCE_MAX_FRAMES and
ETHTOOL_COALESCE_MAX_FRAMES_IRQ.

He also corrects issues such as memory leaks, NULL pointer
dereferences, etc in various error paths. Finally, he corrects debug
prints reporting an unknown state.

The following are changes since commit e45955766b4300e7bbeeaa1c31e0001fe16383e7:
  Merge git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue 40GbE

Przemyslaw Patynowski (7):
  iavf: Fix VLAN_V2 addition/rejection
  iavf: Disallow changing rx/tx-frames and rx/tx-frames-irq
  iavf: Fix reset error handling
  iavf: Fix NULL pointer dereference in iavf_get_link_ksettings
  iavf: Fix adminq error handling
  iavf: Fix handling of dummy receive descriptors
  iavf: Fix missing state logs

 drivers/net/ethernet/intel/iavf/iavf.h        | 14 +++-
 drivers/net/ethernet/intel/iavf/iavf_adminq.c | 15 ++++-
 .../net/ethernet/intel/iavf/iavf_ethtool.c    | 16 ++---
 drivers/net/ethernet/intel/iavf/iavf_main.c   | 22 +++----
 drivers/net/ethernet/intel/iavf/iavf_txrx.c   |  7 +-
 .../net/ethernet/intel/iavf/iavf_virtchnl.c   | 65 ++++++++++++++++++-
 6 files changed, 107 insertions(+), 32 deletions(-)

-- 
2.35.1

[PATCH net 1/7] iavf: Fix VLAN_V2 addition/rejection

[Tony Nguyen]

From: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>

Fix VLAN addition, so that PF driver does not reject whole VLAN batch.
Add VLAN reject handling, so rejected VLANs, won't litter VLAN filter
list. Fix handling of active_(c/s)vlans, so it will be possible to
re-add VLAN filters for user.
Without this patch, after changing trust to off, with VLAN filters
saturated, no VLAN is added, due to PF rejecting addition.

Fixes: 92fc50859872 ("iavf: Restrict maximum VLAN filters for VIRTCHNL_VF_OFFLOAD_VLAN_V2")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Jedrzej Jagielski <jedrzej.jagielski@intel.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf.h        |  9 ++-
 drivers/net/ethernet/intel/iavf/iavf_main.c   | 10 ++-
 .../net/ethernet/intel/iavf/iavf_virtchnl.c   | 65 ++++++++++++++++++-
 3 files changed, 74 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf.h b/drivers/net/ethernet/intel/iavf/iavf.h
index 49aed3e506a6..86bc61c300a7 100644
--- a/drivers/net/ethernet/intel/iavf/iavf.h
+++ b/drivers/net/ethernet/intel/iavf/iavf.h
@@ -159,8 +159,12 @@ struct iavf_vlan {
 struct iavf_vlan_filter {
 	struct list_head list;
 	struct iavf_vlan vlan;
-	bool remove;		/* filter needs to be removed */
-	bool add;		/* filter needs to be added */
+	struct {
+		u8 is_new_vlan:1;	/* filter is new, wait for PF answer */
+		u8 remove:1;		/* filter needs to be removed */
+		u8 add:1;		/* filter needs to be added */
+		u8 padding:5;
+	};
 };
 
 #define IAVF_MAX_TRAFFIC_CLASS	4
@@ -520,6 +524,7 @@ int iavf_get_vf_config(struct iavf_adapter *adapter);
 int iavf_get_vf_vlan_v2_caps(struct iavf_adapter *adapter);
 int iavf_send_vf_offload_vlan_v2_msg(struct iavf_adapter *adapter);
 void iavf_set_queue_vlan_tag_loc(struct iavf_adapter *adapter);
+u16 iavf_get_num_vlans_added(struct iavf_adapter *adapter);
 void iavf_irq_enable(struct iavf_adapter *adapter, bool flush);
 void iavf_configure_queues(struct iavf_adapter *adapter);
 void iavf_deconfigure_queues(struct iavf_adapter *adapter);
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index f3ecb3bca33d..2a8643e66331 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -843,7 +843,7 @@ static void iavf_restore_filters(struct iavf_adapter *adapter)
  * iavf_get_num_vlans_added - get number of VLANs added
  * @adapter: board private structure
  */
-static u16 iavf_get_num_vlans_added(struct iavf_adapter *adapter)
+u16 iavf_get_num_vlans_added(struct iavf_adapter *adapter)
 {
 	return bitmap_weight(adapter->vsi.active_cvlans, VLAN_N_VID) +
 		bitmap_weight(adapter->vsi.active_svlans, VLAN_N_VID);
@@ -906,11 +906,6 @@ static int iavf_vlan_rx_add_vid(struct net_device *netdev,
 	if (!iavf_add_vlan(adapter, IAVF_VLAN(vid, be16_to_cpu(proto))))
 		return -ENOMEM;
 
-	if (proto == cpu_to_be16(ETH_P_8021Q))
-		set_bit(vid, adapter->vsi.active_cvlans);
-	else
-		set_bit(vid, adapter->vsi.active_svlans);
-
 	return 0;
 }
 
@@ -2956,6 +2951,9 @@ static void iavf_reset_task(struct work_struct *work)
 	adapter->aq_required |= IAVF_FLAG_AQ_ADD_CLOUD_FILTER;
 	iavf_misc_irq_enable(adapter);
 
+	bitmap_clear(adapter->vsi.active_cvlans, 0, VLAN_N_VID);
+	bitmap_clear(adapter->vsi.active_svlans, 0, VLAN_N_VID);
+
 	mod_delayed_work(iavf_wq, &adapter->watchdog_task, 2);
 
 	/* We were running when the reset started, so we need to restore some
diff --git a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c
index 782450d5c12f..1603e99bae4a 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c
@@ -626,6 +626,33 @@ static void iavf_mac_add_reject(struct iavf_adapter *adapter)
 	spin_unlock_bh(&adapter->mac_vlan_list_lock);
 }
 
+/**
+ * iavf_vlan_add_reject
+ * @adapter: adapter structure
+ *
+ * Remove VLAN filters from list based on PF response.
+ **/
+static void iavf_vlan_add_reject(struct iavf_adapter *adapter)
+{
+	struct iavf_vlan_filter *f, *ftmp;
+
+	spin_lock_bh(&adapter->mac_vlan_list_lock);
+	list_for_each_entry_safe(f, ftmp, &adapter->vlan_filter_list, list) {
+		if (f->is_new_vlan) {
+			if (f->vlan.tpid == ETH_P_8021Q)
+				clear_bit(f->vlan.vid,
+					  adapter->vsi.active_cvlans);
+			else
+				clear_bit(f->vlan.vid,
+					  adapter->vsi.active_svlans);
+
+			list_del(&f->list);
+			kfree(f);
+		}
+	}
+	spin_unlock_bh(&adapter->mac_vlan_list_lock);
+}
+
 /**
  * iavf_add_vlans
  * @adapter: adapter structure
@@ -683,6 +710,7 @@ void iavf_add_vlans(struct iavf_adapter *adapter)
 				vvfl->vlan_id[i] = f->vlan.vid;
 				i++;
 				f->add = false;
+				f->is_new_vlan = true;
 				if (i == count)
 					break;
 			}
@@ -695,10 +723,18 @@ void iavf_add_vlans(struct iavf_adapter *adapter)
 		iavf_send_pf_msg(adapter, VIRTCHNL_OP_ADD_VLAN, (u8 *)vvfl, len);
 		kfree(vvfl);
 	} else {
+		u16 max_vlans = adapter->vlan_v2_caps.filtering.max_filters;
+		u16 current_vlans = iavf_get_num_vlans_added(adapter);
 		struct virtchnl_vlan_filter_list_v2 *vvfl_v2;
 
 		adapter->current_op = VIRTCHNL_OP_ADD_VLAN_V2;
 
+		if ((count + current_vlans) > max_vlans &&
+		    current_vlans < max_vlans) {
+			count = max_vlans - iavf_get_num_vlans_added(adapter);
+			more = true;
+		}
+
 		len = sizeof(*vvfl_v2) + ((count - 1) *
 					  sizeof(struct virtchnl_vlan_filter));
 		if (len > IAVF_MAX_AQ_BUF_SIZE) {
@@ -725,6 +761,9 @@ void iavf_add_vlans(struct iavf_adapter *adapter)
 					&adapter->vlan_v2_caps.filtering.filtering_support;
 				struct virtchnl_vlan *vlan;
 
+				if (i == count)
+					break;
+
 				/* give priority over outer if it's enabled */
 				if (filtering_support->outer)
 					vlan = &vvfl_v2->filters[i].outer;
@@ -736,8 +775,7 @@ void iavf_add_vlans(struct iavf_adapter *adapter)
 
 				i++;
 				f->add = false;
-				if (i == count)
-					break;
+				f->is_new_vlan = true;
 			}
 		}
 
@@ -2080,6 +2118,11 @@ void iavf_virtchnl_completion(struct iavf_adapter *adapter,
 			 */
 			iavf_netdev_features_vlan_strip_set(netdev, true);
 			break;
+		case VIRTCHNL_OP_ADD_VLAN_V2:
+			iavf_vlan_add_reject(adapter);
+			dev_warn(&adapter->pdev->dev, "Failed to add VLAN filter, error %s\n",
+				 iavf_stat_str(&adapter->hw, v_retval));
+			break;
 		default:
 			dev_err(&adapter->pdev->dev, "PF returned error %d (%s) to our request %d\n",
 				v_retval, iavf_stat_str(&adapter->hw, v_retval),
@@ -2332,6 +2375,24 @@ void iavf_virtchnl_completion(struct iavf_adapter *adapter,
 		spin_unlock_bh(&adapter->adv_rss_lock);
 		}
 		break;
+	case VIRTCHNL_OP_ADD_VLAN_V2: {
+		struct iavf_vlan_filter *f;
+
+		spin_lock_bh(&adapter->mac_vlan_list_lock);
+		list_for_each_entry(f, &adapter->vlan_filter_list, list) {
+			if (f->is_new_vlan) {
+				f->is_new_vlan = false;
+				if (f->vlan.tpid == ETH_P_8021Q)
+					set_bit(f->vlan.vid,
+						adapter->vsi.active_cvlans);
+				else
+					set_bit(f->vlan.vid,
+						adapter->vsi.active_svlans);
+			}
+		}
+		spin_unlock_bh(&adapter->mac_vlan_list_lock);
+		}
+		break;
 	case VIRTCHNL_OP_ENABLE_VLAN_STRIPPING:
 		/* PF enabled vlan strip on this VF.
 		 * Update netdev->features if needed to be in sync with ethtool.
-- 
2.35.1

[PATCH net 2/7] iavf: Disallow changing rx/tx-frames and rx/tx-frames-irq

[Tony Nguyen]

From: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>

Remove from supported_coalesce_params ETHTOOL_COALESCE_MAX_FRAMES
and ETHTOOL_COALESCE_MAX_FRAMES_IRQ. As tx-frames-irq allowed
user to change budget for iavf_clean_tx_irq, remove work_limit
and use define for budget.

Without this patch there would be possibility to change rx/tx-frames
and rx/tx-frames-irq, which for rx/tx-frames did nothing, while for
rx/tx-frames-irq it changed rx/tx-frames and only changed budget
for cleaning NAPI poll.

Fixes: fbb7ddfef253 ("i40evf: core ethtool functionality")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Jun Zhang <xuejun.zhang@intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf.h         |  1 -
 drivers/net/ethernet/intel/iavf/iavf_ethtool.c | 10 ----------
 drivers/net/ethernet/intel/iavf/iavf_main.c    |  1 -
 drivers/net/ethernet/intel/iavf/iavf_txrx.c    |  2 +-
 4 files changed, 1 insertion(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf.h b/drivers/net/ethernet/intel/iavf/iavf.h
index 86bc61c300a7..2a7b3c085aa9 100644
--- a/drivers/net/ethernet/intel/iavf/iavf.h
+++ b/drivers/net/ethernet/intel/iavf/iavf.h
@@ -64,7 +64,6 @@ struct iavf_vsi {
 	u16 id;
 	DECLARE_BITMAP(state, __IAVF_VSI_STATE_SIZE__);
 	int base_vector;
-	u16 work_limit;
 	u16 qs_handle;
 	void *priv;     /* client driver data reference. */
 };
diff --git a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c
index 3bb56714beb0..e535d4c3da49 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c
@@ -692,12 +692,8 @@ static int __iavf_get_coalesce(struct net_device *netdev,
 			       struct ethtool_coalesce *ec, int queue)
 {
 	struct iavf_adapter *adapter = netdev_priv(netdev);
-	struct iavf_vsi *vsi = &adapter->vsi;
 	struct iavf_ring *rx_ring, *tx_ring;
 
-	ec->tx_max_coalesced_frames = vsi->work_limit;
-	ec->rx_max_coalesced_frames = vsi->work_limit;
-
 	/* Rx and Tx usecs per queue value. If user doesn't specify the
 	 * queue, return queue 0's value to represent.
 	 */
@@ -825,12 +821,8 @@ static int __iavf_set_coalesce(struct net_device *netdev,
 			       struct ethtool_coalesce *ec, int queue)
 {
 	struct iavf_adapter *adapter = netdev_priv(netdev);
-	struct iavf_vsi *vsi = &adapter->vsi;
 	int i;
 
-	if (ec->tx_max_coalesced_frames_irq || ec->rx_max_coalesced_frames_irq)
-		vsi->work_limit = ec->tx_max_coalesced_frames_irq;
-
 	if (ec->rx_coalesce_usecs == 0) {
 		if (ec->use_adaptive_rx_coalesce)
 			netif_info(adapter, drv, netdev, "rx-usecs=0, need to disable adaptive-rx for a complete disable\n");
@@ -1969,8 +1961,6 @@ static int iavf_set_rxfh(struct net_device *netdev, const u32 *indir,
 
 static const struct ethtool_ops iavf_ethtool_ops = {
 	.supported_coalesce_params = ETHTOOL_COALESCE_USECS |
-				     ETHTOOL_COALESCE_MAX_FRAMES |
-				     ETHTOOL_COALESCE_MAX_FRAMES_IRQ |
 				     ETHTOOL_COALESCE_USE_ADAPTIVE,
 	.get_drvinfo		= iavf_get_drvinfo,
 	.get_link		= ethtool_op_get_link,
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index 2a8643e66331..2e2c153ce46a 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -2240,7 +2240,6 @@ int iavf_parse_vf_resource_msg(struct iavf_adapter *adapter)
 
 	adapter->vsi.back = adapter;
 	adapter->vsi.base_vector = 1;
-	adapter->vsi.work_limit = IAVF_DEFAULT_IRQ_WORK;
 	vsi->netdev = adapter->netdev;
 	vsi->qs_handle = adapter->vsi_res->qset_handle;
 	if (adapter->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF) {
diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
index 978f651c6b09..7bf8c25dc824 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
@@ -194,7 +194,7 @@ static bool iavf_clean_tx_irq(struct iavf_vsi *vsi,
 	struct iavf_tx_buffer *tx_buf;
 	struct iavf_tx_desc *tx_desc;
 	unsigned int total_bytes = 0, total_packets = 0;
-	unsigned int budget = vsi->work_limit;
+	unsigned int budget = IAVF_DEFAULT_IRQ_WORK;
 
 	tx_buf = &tx_ring->tx_bi[i];
 	tx_desc = IAVF_TX_DESC(tx_ring, i);
-- 
2.35.1

[PATCH net 3/7] iavf: Fix reset error handling

[Tony Nguyen]

From: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>

Do not call iavf_close in iavf_reset_task error handling. Doing so can
lead to double call of napi_disable, which can lead to deadlock there.
Removing VF would lead to iavf_remove task being stuck, because it
requires crit_lock, which is held by iavf_close.
Call iavf_disable_vf if reset fail, so that driver will clean up
remaining invalid resources.
During rapid VF resets, HW can fail to setup VF mailbox. Wrong
error handling can lead to iavf_remove being stuck with:
[ 5218.999087] iavf 0000:82:01.0: Failed to init adminq: -53
...
[ 5267.189211] INFO: task repro.sh:11219 blocked for more than 30 seconds.
[ 5267.189520]       Tainted: G S          E     5.18.0-04958-ga54ce3703613-dirty #1
[ 5267.189764] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 5267.190062] task:repro.sh        state:D stack:    0 pid:11219 ppid:  8162 flags:0x00000000
[ 5267.190347] Call Trace:
[ 5267.190647]  <TASK>
[ 5267.190927]  __schedule+0x460/0x9f0
[ 5267.191264]  schedule+0x44/0xb0
[ 5267.191563]  schedule_preempt_disabled+0x14/0x20
[ 5267.191890]  __mutex_lock.isra.12+0x6e3/0xac0
[ 5267.192237]  ? iavf_remove+0xf9/0x6c0 [iavf]
[ 5267.192565]  iavf_remove+0x12a/0x6c0 [iavf]
[ 5267.192911]  ? _raw_spin_unlock_irqrestore+0x1e/0x40
[ 5267.193285]  pci_device_remove+0x36/0xb0
[ 5267.193619]  device_release_driver_internal+0xc1/0x150
[ 5267.193974]  pci_stop_bus_device+0x69/0x90
[ 5267.194361]  pci_stop_and_remove_bus_device+0xe/0x20
[ 5267.194735]  pci_iov_remove_virtfn+0xba/0x120
[ 5267.195130]  sriov_disable+0x2f/0xe0
[ 5267.195506]  ice_free_vfs+0x7d/0x2f0 [ice]
[ 5267.196056]  ? pci_get_device+0x4f/0x70
[ 5267.196496]  ice_sriov_configure+0x78/0x1a0 [ice]
[ 5267.196995]  sriov_numvfs_store+0xfe/0x140
[ 5267.197466]  kernfs_fop_write_iter+0x12e/0x1c0
[ 5267.197918]  new_sync_write+0x10c/0x190
[ 5267.198404]  vfs_write+0x24e/0x2d0
[ 5267.198886]  ksys_write+0x5c/0xd0
[ 5267.199367]  do_syscall_64+0x3a/0x80
[ 5267.199827]  entry_SYSCALL_64_after_hwframe+0x46/0xb0
[ 5267.200317] RIP: 0033:0x7f5b381205c8
[ 5267.200814] RSP: 002b:00007fff8c7e8c78 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
[ 5267.201981] RAX: ffffffffffffffda RBX: 0000000000000002 RCX: 00007f5b381205c8
[ 5267.202620] RDX: 0000000000000002 RSI: 00005569420ee900 RDI: 0000000000000001
[ 5267.203426] RBP: 00005569420ee900 R08: 000000000000000a R09: 00007f5b38180820
[ 5267.204327] R10: 000000000000000a R11: 0000000000000246 R12: 00007f5b383c06e0
[ 5267.205193] R13: 0000000000000002 R14: 00007f5b383bb880 R15: 0000000000000002
[ 5267.206041]  </TASK>
[ 5267.206970] Kernel panic - not syncing: hung_task: blocked tasks
[ 5267.207809] CPU: 48 PID: 551 Comm: khungtaskd Kdump: loaded Tainted: G S          E     5.18.0-04958-ga54ce3703613-dirty #1
[ 5267.208726] Hardware name: Dell Inc. PowerEdge R730/0WCJNT, BIOS 2.11.0 11/02/2019
[ 5267.209623] Call Trace:
[ 5267.210569]  <TASK>
[ 5267.211480]  dump_stack_lvl+0x33/0x42
[ 5267.212472]  panic+0x107/0x294
[ 5267.213467]  watchdog.cold.8+0xc/0xbb
[ 5267.214413]  ? proc_dohung_task_timeout_secs+0x30/0x30
[ 5267.215511]  kthread+0xf4/0x120
[ 5267.216459]  ? kthread_complete_and_exit+0x20/0x20
[ 5267.217505]  ret_from_fork+0x22/0x30
[ 5267.218459]  </TASK>

Fixes: f0db78928783 ("i40evf: use netdev variable in reset task")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Mateusz Palczewski <mateusz.palczewski@intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf_main.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index 2e2c153ce46a..d200835e795c 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -2998,12 +2998,15 @@ static void iavf_reset_task(struct work_struct *work)
 
 	return;
 reset_err:
+	if (running) {
+		set_bit(__IAVF_VSI_DOWN, adapter->vsi.state);
+		iavf_free_traffic_irqs(adapter);
+	}
+	iavf_disable_vf(adapter);
+
 	mutex_unlock(&adapter->client_lock);
 	mutex_unlock(&adapter->crit_lock);
-	if (running)
-		iavf_change_state(adapter, __IAVF_RUNNING);
 	dev_err(&adapter->pdev->dev, "failed to allocate resources during reinit\n");
-	iavf_close(netdev);
 }
 
 /**
-- 
2.35.1

[PATCH net 4/7] iavf: Fix NULL pointer dereference in iavf_get_link_ksettings

[Tony Nguyen]

From: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>

Fix possible NULL pointer dereference, due to freeing of adapter->vf_res
in iavf_init_get_resources. Commit 209f2f9c7181 ("iavf: Add support for
VIRTCHNL_VF_OFFLOAD_VLAN_V2 negotiation") introduced a regression
where receiving IAVF_ERR_ADMIN_QUEUE_NO_WORK from iavf_get_vf_config
would free adapter->vf_res. However, netdev is still registered, so
ethtool_ops can be called. Calling iavf_get_link_ksettings with no vf_res,
will result with:
[ 9385.242676] BUG: kernel NULL pointer dereference, address: 0000000000000008
[ 9385.242683] #PF: supervisor read access in kernel mode
[ 9385.242686] #PF: error_code(0x0000) - not-present page
[ 9385.242690] PGD 0 P4D 0
[ 9385.242696] Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC PTI
[ 9385.242701] CPU: 6 PID: 3217 Comm: pmdalinux Kdump: loaded Tainted: G S          E     5.18.0-04958-ga54ce3703613-dirty #1
[ 9385.242708] Hardware name: Dell Inc. PowerEdge R730/0WCJNT, BIOS 2.11.0 11/02/2019
[ 9385.242710] RIP: 0010:iavf_get_link_ksettings+0x29/0xd0 [iavf]
[ 9385.242745] Code: 00 0f 1f 44 00 00 b8 01 ef ff ff 48 c7 46 30 00 00 00 00 48 c7 46 38 00 00 00 00 c6 46 0b 00 66 89 46 08 48 8b 87 68 0e 00 00 <f6> 40 08 80 75 50 8b 87 5c 0e 00 00 83 f8 08 74 7a 76 1d 83 f8 20
[ 9385.242749] RSP: 0018:ffffc0560ec7fbd0 EFLAGS: 00010246
[ 9385.242755] RAX: 0000000000000000 RBX: ffffc0560ec7fc08 RCX: 0000000000000000
[ 9385.242759] RDX: ffffffffc0ad4550 RSI: ffffc0560ec7fc08 RDI: ffffa0fc66674000
[ 9385.242762] RBP: 00007ffd1fb2bf50 R08: b6a2d54b892363ee R09: ffffa101dc14fb00
[ 9385.242765] R10: 0000000000000000 R11: 0000000000000004 R12: ffffa0fc66674000
[ 9385.242768] R13: 0000000000000000 R14: ffffa0fc66674000 R15: 00000000ffffffa1
[ 9385.242771] FS:  00007f93711a2980(0000) GS:ffffa0fad72c0000(0000) knlGS:0000000000000000
[ 9385.242775] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 9385.242778] CR2: 0000000000000008 CR3: 0000000a8e61c003 CR4: 00000000003706e0
[ 9385.242781] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 9385.242784] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 9385.242787] Call Trace:
[ 9385.242791]  <TASK>
[ 9385.242793]  ethtool_get_settings+0x71/0x1a0
[ 9385.242814]  __dev_ethtool+0x426/0x2f40
[ 9385.242823]  ? slab_post_alloc_hook+0x4f/0x280
[ 9385.242836]  ? kmem_cache_alloc_trace+0x15d/0x2f0
[ 9385.242841]  ? dev_ethtool+0x59/0x170
[ 9385.242848]  dev_ethtool+0xa7/0x170
[ 9385.242856]  dev_ioctl+0xc3/0x520
[ 9385.242866]  sock_do_ioctl+0xa0/0xe0
[ 9385.242877]  sock_ioctl+0x22f/0x320
[ 9385.242885]  __x64_sys_ioctl+0x84/0xc0
[ 9385.242896]  do_syscall_64+0x3a/0x80
[ 9385.242904]  entry_SYSCALL_64_after_hwframe+0x46/0xb0
[ 9385.242918] RIP: 0033:0x7f93702396db
[ 9385.242923] Code: 73 01 c3 48 8b 0d ad 57 38 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 7d 57 38 00 f7 d8 64 89 01 48
[ 9385.242927] RSP: 002b:00007ffd1fb2bf18 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[ 9385.242932] RAX: ffffffffffffffda RBX: 000055671b1d2fe0 RCX: 00007f93702396db
[ 9385.242935] RDX: 00007ffd1fb2bf20 RSI: 0000000000008946 RDI: 0000000000000007
[ 9385.242937] RBP: 00007ffd1fb2bf20 R08: 0000000000000003 R09: 0030763066307330
[ 9385.242940] R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffd1fb2bf80
[ 9385.242942] R13: 0000000000000007 R14: 0000556719f6de90 R15: 00007ffd1fb2c1b0
[ 9385.242948]  </TASK>
[ 9385.242949] Modules linked in: iavf(E) xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nft_compat nf_nat_tftp nft_objref nf_conntrack_tftp bridge stp llc nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip_set nf_tables rfkill nfnetlink vfat fat irdma ib_uverbs ib_core intel_rapl_msr intel_rapl_common sb_edac x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm iTCO_wdt iTCO_vendor_support ice irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel rapl i40e pcspkr intel_cstate joydev mei_me intel_uncore mxm_wmi mei ipmi_ssif lpc_ich ipmi_si acpi_power_meter xfs libcrc32c mgag200 i2c_algo_bit drm_shmem_helper drm_kms_helper sd_mod t10_pi crc64_rocksoft crc64 syscopyarea sg sysfillrect sysimgblt fb_sys_fops drm ixgbe ahci libahci libata crc32c_intel mdio dca wmi dm_mirror dm_region_hash dm_log dm_mod ipmi_devintf ipmi_msghandler fuse
[ 9385.243065]  [last unloaded: iavf]

Dereference happens in if (ADV_LINK_SUPPORT(adapter)) statement

Fixes: 209f2f9c7181 ("iavf: Add support for VIRTCHNL_VF_OFFLOAD_VLAN_V2 negotiation")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Mateusz Palczewski <mateusz.palczewski@intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf_ethtool.c | 6 ++++++
 drivers/net/ethernet/intel/iavf/iavf_main.c    | 2 +-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c
index e535d4c3da49..01e4037407e5 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c
@@ -275,6 +275,12 @@ static int iavf_get_link_ksettings(struct net_device *netdev,
 {
 	struct iavf_adapter *adapter = netdev_priv(netdev);
 
+	/* if VF failed initialization during __IAVF_INIT_GET_RESOURCES,
+	 * it is possible for this variable to be freed there.
+	 */
+	if (!adapter->vf_res)
+		return -EAGAIN;
+
 	ethtool_link_ksettings_zero_link_mode(cmd, supported);
 	cmd->base.autoneg = AUTONEG_DISABLE;
 	cmd->base.port = PORT_NONE;
diff --git a/drivers/net/ethernet/intel/iavf/iavf_main.c b/drivers/net/ethernet/intel/iavf/iavf_main.c
index d200835e795c..7b3f2f1cc3aa 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_main.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_main.c
@@ -2281,7 +2281,7 @@ static void iavf_init_get_resources(struct iavf_adapter *adapter)
 	err = iavf_get_vf_config(adapter);
 	if (err == -EALREADY) {
 		err = iavf_send_vf_config_msg(adapter);
-		goto err_alloc;
+		goto err;
 	} else if (err == -EINVAL) {
 		/* We only get -EINVAL if the device is in a very bad
 		 * state or if we've been disabled for previous bad
-- 
2.35.1

[PATCH net 5/7] iavf: Fix adminq error handling

[Tony Nguyen]

From: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>

iavf_alloc_asq_bufs/iavf_alloc_arq_bufs allocates with dma_alloc_coherent
memory for VF mailbox.
Free DMA regions for both ASQ and ARQ in case error happens during
configuration of ASQ/ARQ registers.
Without this change it is possible to see when unloading interface:
74626.583369: dma_debug_device_change: device driver has pending DMA allocations while released from device [count=32]
One of leaked entries details: [device address=0x0000000b27ff9000] [size=4096 bytes] [mapped with DMA_BIDIRECTIONAL] [mapped as coherent]

Fixes: d358aa9a7a2d ("i40evf: init code and hardware support")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Mateusz Palczewski <mateusz.palczewski@intel.com>
Tested-by: Marek Szlosek <marek.szlosek@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf_adminq.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf_adminq.c b/drivers/net/ethernet/intel/iavf/iavf_adminq.c
index cd4e6a22d0f9..9ffbd24d83cb 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_adminq.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_adminq.c
@@ -324,6 +324,7 @@ static enum iavf_status iavf_config_arq_regs(struct iavf_hw *hw)
 static enum iavf_status iavf_init_asq(struct iavf_hw *hw)
 {
 	enum iavf_status ret_code = 0;
+	int i;
 
 	if (hw->aq.asq.count > 0) {
 		/* queue already initialized */
@@ -354,12 +355,17 @@ static enum iavf_status iavf_init_asq(struct iavf_hw *hw)
 	/* initialize base registers */
 	ret_code = iavf_config_asq_regs(hw);
 	if (ret_code)
-		goto init_adminq_free_rings;
+		goto init_free_asq_bufs;
 
 	/* success! */
 	hw->aq.asq.count = hw->aq.num_asq_entries;
 	goto init_adminq_exit;
 
+init_free_asq_bufs:
+	for (i = 0; i < hw->aq.num_asq_entries; i++)
+		iavf_free_dma_mem(hw, &hw->aq.asq.r.asq_bi[i]);
+	iavf_free_virt_mem(hw, &hw->aq.asq.dma_head);
+
 init_adminq_free_rings:
 	iavf_free_adminq_asq(hw);
 
@@ -383,6 +389,7 @@ static enum iavf_status iavf_init_asq(struct iavf_hw *hw)
 static enum iavf_status iavf_init_arq(struct iavf_hw *hw)
 {
 	enum iavf_status ret_code = 0;
+	int i;
 
 	if (hw->aq.arq.count > 0) {
 		/* queue already initialized */
@@ -413,12 +420,16 @@ static enum iavf_status iavf_init_arq(struct iavf_hw *hw)
 	/* initialize base registers */
 	ret_code = iavf_config_arq_regs(hw);
 	if (ret_code)
-		goto init_adminq_free_rings;
+		goto init_free_arq_bufs;
 
 	/* success! */
 	hw->aq.arq.count = hw->aq.num_arq_entries;
 	goto init_adminq_exit;
 
+init_free_arq_bufs:
+	for (i = 0; i < hw->aq.num_arq_entries; i++)
+		iavf_free_dma_mem(hw, &hw->aq.arq.r.arq_bi[i]);
+	iavf_free_virt_mem(hw, &hw->aq.arq.dma_head);
 init_adminq_free_rings:
 	iavf_free_adminq_arq(hw);
 
-- 
2.35.1

[PATCH net 6/7] iavf: Fix handling of dummy receive descriptors

[Tony Nguyen]

From: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>

Fix memory leak caused by not handling dummy receive descriptor properly.
iavf_get_rx_buffer now sets the rx_buffer return value for dummy receive
descriptors. Without this patch, when the hardware writes a dummy
descriptor, iavf would not free the page allocated for the previous receive
buffer. This is an unlikely event but can still happen.

[Jesse: massaged commit message]

Fixes: efa14c398582 ("iavf: allow null RX descriptors")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf_txrx.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
index 7bf8c25dc824..06d18797d25a 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
@@ -1285,11 +1285,10 @@ static struct iavf_rx_buffer *iavf_get_rx_buffer(struct iavf_ring *rx_ring,
 {
 	struct iavf_rx_buffer *rx_buffer;
 
-	if (!size)
-		return NULL;
-
 	rx_buffer = &rx_ring->rx_bi[rx_ring->next_to_clean];
 	prefetchw(rx_buffer->page);
+	if (!size)
+		return rx_buffer;
 
 	/* we are reusing so sync this buffer for CPU use */
 	dma_sync_single_range_for_cpu(rx_ring->dev,
-- 
2.35.1

[PATCH net 7/7] iavf: Fix missing state logs

[Tony Nguyen]

From: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>

Fix debug prints, by adding missing state prints.

Extend iavf_state_str by strings for __IAVF_INIT_EXTENDED_CAPS and
__IAVF_INIT_CONFIG_ADAPTER.

Without this patch, when enabling debug prints for iavf.h, user will
see:
iavf 0000:06:0e.0: state transition from:__IAVF_INIT_GET_RESOURCES to:__IAVF_UNKNOWN_STATE
iavf 0000:06:0e.0: state transition from:__IAVF_UNKNOWN_STATE to:__IAVF_UNKNOWN_STATE

Fixes: 605ca7c5c670 ("iavf: Fix kernel BUG in free_msi_irqs")
Signed-off-by: Przemyslaw Patynowski <przemyslawx.patynowski@intel.com>
Signed-off-by: Jun Zhang <xuejun.zhang@intel.com>
Tested-by: Konrad Jankowski <konrad0.jankowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
---
 drivers/net/ethernet/intel/iavf/iavf.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/intel/iavf/iavf.h b/drivers/net/ethernet/intel/iavf/iavf.h
index 2a7b3c085aa9..0ea0361cd86b 100644
--- a/drivers/net/ethernet/intel/iavf/iavf.h
+++ b/drivers/net/ethernet/intel/iavf/iavf.h
@@ -464,6 +464,10 @@ static inline const char *iavf_state_str(enum iavf_state_t state)
 		return "__IAVF_INIT_VERSION_CHECK";
 	case __IAVF_INIT_GET_RESOURCES:
 		return "__IAVF_INIT_GET_RESOURCES";
+	case __IAVF_INIT_EXTENDED_CAPS:
+		return "__IAVF_INIT_EXTENDED_CAPS";
+	case __IAVF_INIT_CONFIG_ADAPTER:
+		return "__IAVF_INIT_CONFIG_ADAPTER";
 	case __IAVF_INIT_SW:
 		return "__IAVF_INIT_SW";
 	case __IAVF_INIT_FAILED:
-- 
2.35.1