[net-next 00/11][pull request] 40GbE Intel Wired LAN Driver Updates 2018-05-14

[net-next 00/11][pull request] 40GbE Intel Wired LAN Driver Updates 2018-05-14

[Jeff Kirsher]

This series contains updates to virtchnl, i40e and i40evf.

Bruce cleans up whitespace and unnecessary parentheses in virtchnl.

Jake does a number of stat cleanups in the i40e driver, including
cleanup of code indentation, whitespace issues, remove duplicate stats,
fix grammar in code comment and general spring cleaning of the
statistics code.

Patryk fixes an issue where we recalculate vectors left and vectors
wanted but do not take into account the reduced number of queue pairs
per VSI.

Harshitha adds tx_busy stat to ethtool stats to track the number of
times we return NETDEV_TX_BUSY to the stack during transmit.

Paweł fixes a potential system crash when unloading the VF driver after
a hardware reset.

The following are changes since commit 289e1f4e9e4a09c73a1c0152bb93855ea351ccda:
  net: ipv4: ipconfig: fix unused variable
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue 40GbE

Bruce Allan (1):
  virtchnl: Whitespace and parenthesis cleanup

Harshitha Ramamurthy (1):
  i40e: add tx_busy to ethtool stats

Jacob Keller (6):
  i40e: calculate ethtool stats size in a separate function
  i40e: remove duplicate pfc stats
  i40e: cleanup whitespace for some ethtool stat definitions
  i40evf: remove MAX_QUEUES and just use I40EVF_MAX_REQ_QUEUES
  i40e: cleanup wording in a header comment
  i40e: free the skb after clearing the bitlock

Jeff Kirsher (1):
  i40evf: Fix client header define

Patryk Małek (1):
  i40e: Fix recalculation of MSI-X vectors for VMDq

Paweł Jabłoński (1):
  i40evf: Fix a hardware reset support in VF driver

 .../net/ethernet/intel/i40e/i40e_ethtool.c    | 45 ++++++++++---------
 drivers/net/ethernet/intel/i40e/i40e_main.c   | 36 ++++++++++++---
 drivers/net/ethernet/intel/i40e/i40e_ptp.c    |  9 +++-
 drivers/net/ethernet/intel/i40evf/i40evf.h    |  1 -
 .../net/ethernet/intel/i40evf/i40evf_client.h |  6 +--
 .../net/ethernet/intel/i40evf/i40evf_main.c   |  8 ++--
 include/linux/avf/virtchnl.h                  |  4 +-
 7 files changed, 72 insertions(+), 37 deletions(-)

-- 
2.17.0

[net-next 01/11] virtchnl: Whitespace and parenthesis cleanup

[Jeff Kirsher]

From: Bruce Allan <bruce.w.allan@intel.com>

Clean up existing instances of unnecessary parentheses in if
statement and change order of conditionals to make it easier to read

The opening /* should be followed by a single space and the closing */
should be preceded with a single space.

Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 include/linux/avf/virtchnl.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/avf/virtchnl.h b/include/linux/avf/virtchnl.h
index b0a7f315bfbe..212b3822d180 100644
--- a/include/linux/avf/virtchnl.h
+++ b/include/linux/avf/virtchnl.h
@@ -485,7 +485,7 @@ VIRTCHNL_CHECK_STRUCT_LEN(6, virtchnl_rss_key);
 struct virtchnl_rss_lut {
 	u16 vsi_id;
 	u16 lut_entries;
-	u8 lut[1];        /* RSS lookup table*/
+	u8 lut[1];        /* RSS lookup table */
 };
 
 VIRTCHNL_CHECK_STRUCT_LEN(6, virtchnl_rss_lut);
@@ -819,7 +819,7 @@ virtchnl_vc_validate_vf_msg(struct virtchnl_version_info *ver, u32 v_opcode,
 		return VIRTCHNL_ERR_PARAM;
 	}
 	/* few more checks */
-	if ((valid_len != msglen) || (err_msg_format))
+	if (err_msg_format || valid_len != msglen)
 		return VIRTCHNL_STATUS_ERR_OPCODE_MISMATCH;
 
 	return 0;
-- 
2.17.0

[net-next 02/11] i40evf: Fix client header define

[Jeff Kirsher]

Fix up the VF client header define, since it is the same as the PF
client header.

Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf_client.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_client.h b/drivers/net/ethernet/intel/i40evf/i40evf_client.h
index fc6592c3de9c..5585f362048a 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_client.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_client.h
@@ -1,8 +1,8 @@
 /* SPDX-License-Identifier: GPL-2.0 */
 /* Copyright(c) 2013 - 2018 Intel Corporation. */
 
-#ifndef _I40E_CLIENT_H_
-#define _I40E_CLIENT_H_
+#ifndef _I40EVF_CLIENT_H_
+#define _I40EVF_CLIENT_H_
 
 #define I40EVF_CLIENT_STR_LENGTH 10
 
@@ -166,4 +166,4 @@ struct i40e_client {
 /* used by clients */
 int i40evf_register_client(struct i40e_client *client);
 int i40evf_unregister_client(struct i40e_client *client);
-#endif /* _I40E_CLIENT_H_ */
+#endif /* _I40EVF_CLIENT_H_ */
-- 
2.17.0

[net-next 03/11] i40e: calculate ethtool stats size in a separate function

[Jeff Kirsher]

From: Jacob Keller <jacob.e.keller@intel.com>

Use a separate function to calculate the number of stats for
a particular device. This helps reduce the clutter in
i40e_get_sset_count().

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 .../net/ethernet/intel/i40e/i40e_ethtool.c    | 28 ++++++++++++-------
 1 file changed, 18 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index fc6a5eef141c..a62142f033d2 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -1658,6 +1658,23 @@ static int i40e_set_ringparam(struct net_device *netdev,
 	return err;
 }
 
+static int i40e_get_stats_count(struct net_device *netdev)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+
+	if (vsi == pf->vsi[pf->lan_vsi] && pf->hw.partition_id == 1) {
+		if (pf->lan_veb != I40E_NO_VEB &&
+		    pf->flags & I40E_FLAG_VEB_STATS_ENABLED)
+			return I40E_PF_STATS_LEN(netdev) + I40E_VEB_STATS_TOTAL;
+		else
+			return I40E_PF_STATS_LEN(netdev);
+	} else {
+		return I40E_VSI_STATS_LEN(netdev);
+	}
+}
+
 static int i40e_get_sset_count(struct net_device *netdev, int sset)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
@@ -1668,16 +1685,7 @@ static int i40e_get_sset_count(struct net_device *netdev, int sset)
 	case ETH_SS_TEST:
 		return I40E_TEST_LEN;
 	case ETH_SS_STATS:
-		if (vsi == pf->vsi[pf->lan_vsi] && pf->hw.partition_id == 1) {
-			int len = I40E_PF_STATS_LEN(netdev);
-
-			if ((pf->lan_veb != I40E_NO_VEB) &&
-			    (pf->flags & I40E_FLAG_VEB_STATS_ENABLED))
-				len += I40E_VEB_STATS_TOTAL;
-			return len;
-		} else {
-			return I40E_VSI_STATS_LEN(netdev);
-		}
+		return i40e_get_stats_count(netdev);
 	case ETH_SS_PRIV_FLAGS:
 		return I40E_PRIV_FLAGS_STR_LEN +
 			(pf->hw.pf_id == 0 ? I40E_GL_PRIV_FLAGS_STR_LEN : 0);
-- 
2.17.0

[net-next 04/11] i40e: remove duplicate pfc stats

[Jeff Kirsher]

From: Jacob Keller <jacob.e.keller@intel.com>

The pfc related priority stats are already handled separately as these
stats are actually arrays of length I40E_MAX_USER_PRIORITY. Thus,
including them within i40e_gstrings_stats will just duplicate data.

Worse, the sizeof will be incorrect, as it will be the total size of the
stat arrays, which in this case is 8 * sizeof(u64), so we will only copy
the stat contents as if they were a u32.

Since we already correctly handle these stats else where, remove them
from the i40e_gstrings_stats.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index a62142f033d2..0f237397f52b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -103,10 +103,6 @@ static const struct i40e_stats i40e_gstrings_stats[] = {
 	I40E_PF_STAT("link_xoff_rx", stats.link_xoff_rx),
 	I40E_PF_STAT("link_xon_tx", stats.link_xon_tx),
 	I40E_PF_STAT("link_xoff_tx", stats.link_xoff_tx),
-	I40E_PF_STAT("priority_xon_rx", stats.priority_xon_rx),
-	I40E_PF_STAT("priority_xoff_rx", stats.priority_xoff_rx),
-	I40E_PF_STAT("priority_xon_tx", stats.priority_xon_tx),
-	I40E_PF_STAT("priority_xoff_tx", stats.priority_xoff_tx),
 	I40E_PF_STAT("rx_size_64", stats.rx_size_64),
 	I40E_PF_STAT("rx_size_127", stats.rx_size_127),
 	I40E_PF_STAT("rx_size_255", stats.rx_size_255),
-- 
2.17.0

[net-next 05/11] i40e: cleanup whitespace for some ethtool stat definitions

[Jeff Kirsher]

From: Jacob Keller <jacob.e.keller@intel.com>

A future patch is going to refactor some of the ethtool statistic code.
To keep the patches easy to review, cleanup some of the indentation used
for macro definitions first.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 0f237397f52b..e6e58e8404c5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -19,13 +19,13 @@ struct i40e_stats {
 }
 
 #define I40E_NETDEV_STAT(_net_stat) \
-		I40E_STAT(struct rtnl_link_stats64, #_net_stat, _net_stat)
+	I40E_STAT(struct rtnl_link_stats64, #_net_stat, _net_stat)
 #define I40E_PF_STAT(_name, _stat) \
-		I40E_STAT(struct i40e_pf, _name, _stat)
+	I40E_STAT(struct i40e_pf, _name, _stat)
 #define I40E_VSI_STAT(_name, _stat) \
-		I40E_STAT(struct i40e_vsi, _name, _stat)
+	I40E_STAT(struct i40e_vsi, _name, _stat)
 #define I40E_VEB_STAT(_name, _stat) \
-		I40E_STAT(struct i40e_veb, _name, _stat)
+	I40E_STAT(struct i40e_veb, _name, _stat)
 
 static const struct i40e_stats i40e_gstrings_net_stats[] = {
 	I40E_NETDEV_STAT(rx_packets),
@@ -144,9 +144,9 @@ static const struct i40e_stats i40e_gstrings_stats[] = {
 	    * 2 /* Tx and Rx together */                                     \
 	    * (sizeof(struct i40e_queue_stats) / sizeof(u64)))
 #define I40E_GLOBAL_STATS_LEN	ARRAY_SIZE(i40e_gstrings_stats)
-#define I40E_NETDEV_STATS_LEN   ARRAY_SIZE(i40e_gstrings_net_stats)
+#define I40E_NETDEV_STATS_LEN	ARRAY_SIZE(i40e_gstrings_net_stats)
 #define I40E_MISC_STATS_LEN	ARRAY_SIZE(i40e_gstrings_misc_stats)
-#define I40E_VSI_STATS_LEN(n)   (I40E_NETDEV_STATS_LEN + \
+#define I40E_VSI_STATS_LEN(n)	(I40E_NETDEV_STATS_LEN + \
 				 I40E_MISC_STATS_LEN + \
 				 I40E_QUEUE_STATS_LEN((n)))
 #define I40E_PFC_STATS_LEN ( \
-- 
2.17.0

[net-next 06/11] i40e: Fix recalculation of MSI-X vectors for VMDq

[Jeff Kirsher]

From: Patryk Małek <patryk.malek@intel.com>

This patch adds a recalculation of number of MSI-X
vectors for VMDq in the case where we have less
vectors available than we would want to reserve for
VMDq.

It fixes the issue where we recalculate vectors left
and vectors wanted but we didn't take into account
the reduced number of queue pairs per VSI.

Signed-off-by: Patryk Małek <patryk.malek@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index c8659fbd7111..f17867ab9a90 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -10309,21 +10309,28 @@ static int i40e_init_msix(struct i40e_pf *pf)
 
 	/* any vectors left over go for VMDq support */
 	if (pf->flags & I40E_FLAG_VMDQ_ENABLED) {
-		int vmdq_vecs_wanted = pf->num_vmdq_vsis * pf->num_vmdq_qps;
-		int vmdq_vecs = min_t(int, vectors_left, vmdq_vecs_wanted);
-
 		if (!vectors_left) {
 			pf->num_vmdq_msix = 0;
 			pf->num_vmdq_qps = 0;
 		} else {
+			int vmdq_vecs_wanted =
+				pf->num_vmdq_vsis * pf->num_vmdq_qps;
+			int vmdq_vecs =
+				min_t(int, vectors_left, vmdq_vecs_wanted);
+
 			/* if we're short on vectors for what's desired, we limit
 			 * the queues per vmdq.  If this is still more than are
 			 * available, the user will need to change the number of
 			 * queues/vectors used by the PF later with the ethtool
 			 * channels command
 			 */
-			if (vmdq_vecs < vmdq_vecs_wanted)
+			if (vectors_left < vmdq_vecs_wanted) {
 				pf->num_vmdq_qps = 1;
+				vmdq_vecs_wanted = pf->num_vmdq_vsis;
+				vmdq_vecs = min_t(int,
+						  vectors_left,
+						  vmdq_vecs_wanted);
+			}
 			pf->num_vmdq_msix = pf->num_vmdq_qps;
 
 			v_budget += vmdq_vecs;
-- 
2.17.0

[net-next 07/11] i40e: add tx_busy to ethtool stats

[Jeff Kirsher]

From: Harshitha Ramamurthy <harshitha.ramamurthy@intel.com>

This patch adds the tx_busy stat to the ethtool stats. The tx_busy
stat tracks the number of times we return NETDEV_TX_BUSY to the stack
during transmit.

Signed-off-by: Harshitha Ramamurthy <harshitha.ramamurthy@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index e6e58e8404c5..329e59eae4a1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -66,6 +66,7 @@ static const struct i40e_stats i40e_gstrings_misc_stats[] = {
 	I40E_VSI_STAT("rx_unknown_protocol", eth_stats.rx_unknown_protocol),
 	I40E_VSI_STAT("tx_linearize", tx_linearize),
 	I40E_VSI_STAT("tx_force_wb", tx_force_wb),
+	I40E_VSI_STAT("tx_busy", tx_busy),
 	I40E_VSI_STAT("rx_alloc_fail", rx_buf_failed),
 	I40E_VSI_STAT("rx_pg_alloc_fail", rx_page_failed),
 };
-- 
2.17.0

[net-next 08/11] i40evf: remove MAX_QUEUES and just use I40EVF_MAX_REQ_QUEUES

[Jeff Kirsher]

From: Jacob Keller <jacob.e.keller@intel.com>

We don't really need to have separate definitions for MAX_QUEUES and
I40EVF_MAX_REQ_QUEUES, since we'll always be limited by how many queues
we request anyways. If we haven't enabled requesting the maximum number
of queues, there's no reason to have our call to alloc_etherdev_mq
actually pass the higher value, since we'd never enable those queues
anyways.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf.h      | 1 -
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 5 +++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index 98b834932dd3..96e537a35000 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -81,7 +81,6 @@ struct i40e_vsi {
 #define I40E_TX_DESC(R, i) (&(((struct i40e_tx_desc *)((R)->desc))[i]))
 #define I40E_TX_CTXTDESC(R, i) \
 	(&(((struct i40e_tx_context_desc *)((R)->desc))[i]))
-#define MAX_QUEUES 16
 #define I40EVF_MAX_REQ_QUEUES 4
 
 #define I40EVF_HKEY_ARRAY_SIZE ((I40E_VFQF_HKEY_MAX_INDEX + 1) * 4)
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 3f04a182903d..95a222d7ae43 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -2331,7 +2331,7 @@ static int i40evf_validate_ch_config(struct i40evf_adapter *adapter,
 		total_max_rate += tx_rate;
 		num_qps += mqprio_qopt->qopt.count[i];
 	}
-	if (num_qps > MAX_QUEUES)
+	if (num_qps > I40EVF_MAX_REQ_QUEUES)
 		return -EINVAL;
 
 	ret = i40evf_validate_tx_bandwidth(adapter, total_max_rate);
@@ -3689,7 +3689,8 @@ static int i40evf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_master(pdev);
 
-	netdev = alloc_etherdev_mq(sizeof(struct i40evf_adapter), MAX_QUEUES);
+	netdev = alloc_etherdev_mq(sizeof(struct i40evf_adapter),
+				   I40EVF_MAX_REQ_QUEUES);
 	if (!netdev) {
 		err = -ENOMEM;
 		goto err_alloc_etherdev;
-- 
2.17.0

[net-next 09/11] i40e: cleanup wording in a header comment

[Jeff Kirsher]

From: Jacob Keller <jacob.e.keller@intel.com>

Fix up the English in the header comment for i40e_ptp_tx_hang.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ptp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ptp.c b/drivers/net/ethernet/intel/i40e/i40e_ptp.c
index aa3daec2049d..6706141a5ccd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ptp.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ptp.c
@@ -317,7 +317,7 @@ void i40e_ptp_rx_hang(struct i40e_pf *pf)
  * This watchdog task is run periodically to make sure that we clear the Tx
  * timestamp logic if we don't obtain a timestamp in a reasonable amount of
  * time. It is unexpected in the normal case but if it occurs it results in
- * permanently prevent timestamps of future packets
+ * permanently preventing timestamps of future packets.
  **/
 void i40e_ptp_tx_hang(struct i40e_pf *pf)
 {
-- 
2.17.0

[net-next 10/11] i40e: free the skb after clearing the bitlock

[Jeff Kirsher]

From: Jacob Keller <jacob.e.keller@intel.com>

In commit bbc4e7d273b5 ("i40e: fix race condition with PTP_TX_IN_PROGRESS
bits") we modified the code which handles Tx timestamps so that we would
clear the progress bit as soon as possible.

A later commit 0bc0706b46cd ("i40e: check for Tx timestamp timeouts during
watchdog") introduced similar code for detecting and handling cleanup of
a blocked Tx timestamp. This code did not use the same pattern for cleaning
up the skb.

Update this code to wait to free the skb until after the bit lock is
free, by first setting the ptp_tx_skb to NULL and clearing the lock.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ptp.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ptp.c b/drivers/net/ethernet/intel/i40e/i40e_ptp.c
index 6706141a5ccd..d50d84927e6b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ptp.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ptp.c
@@ -321,6 +321,8 @@ void i40e_ptp_rx_hang(struct i40e_pf *pf)
  **/
 void i40e_ptp_tx_hang(struct i40e_pf *pf)
 {
+	struct sk_buff *skb;
+
 	if (!(pf->flags & I40E_FLAG_PTP) || !pf->ptp_tx)
 		return;
 
@@ -333,9 +335,12 @@ void i40e_ptp_tx_hang(struct i40e_pf *pf)
 	 * within a second it is reasonable to assume that we never will.
 	 */
 	if (time_is_before_jiffies(pf->ptp_tx_start + HZ)) {
-		dev_kfree_skb_any(pf->ptp_tx_skb);
+		skb = pf->ptp_tx_skb;
 		pf->ptp_tx_skb = NULL;
 		clear_bit_unlock(__I40E_PTP_TX_IN_PROGRESS, pf->state);
+
+		/* Free the skb after we clear the bitlock */
+		dev_kfree_skb_any(skb);
 		pf->tx_hwtstamp_timeouts++;
 	}
 }
-- 
2.17.0

[net-next 11/11] i40evf: Fix a hardware reset support in VF driver

[Jeff Kirsher]

From: Paweł Jabłoński <pawel.jablonski@intel.com>

This patch fixes a hardware reset support in VF driver.
It is needed because when a hardware reset is detected
adapter->state is in __I40EVF_RESETTING state before
i40evf_reset_task is called. Without this patch
unloading VF driver after a hardware reset ends
with a system crash.

Signed-off-by: Paweł Jabłoński <pawel.jablonski@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c   | 21 +++++++++++++++++--
 .../net/ethernet/intel/i40evf/i40evf_main.c   |  3 ++-
 2 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index f17867ab9a90..b5daa5c9c7de 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2840,6 +2840,23 @@ static int i40e_vlan_rx_add_vid(struct net_device *netdev,
 	return ret;
 }
 
+/**
+ * i40e_vlan_rx_add_vid_up - Add a vlan id filter to HW offload in UP path
+ * @netdev: network interface to be adjusted
+ * @proto: unused protocol value
+ * @vid: vlan id to be added
+ **/
+static void i40e_vlan_rx_add_vid_up(struct net_device *netdev,
+				    __always_unused __be16 proto, u16 vid)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+
+	if (vid >= VLAN_N_VID)
+		return;
+	set_bit(vid, vsi->active_vlans);
+}
+
 /**
  * i40e_vlan_rx_kill_vid - Remove a vlan id filter from HW offload
  * @netdev: network interface to be adjusted
@@ -2882,8 +2899,8 @@ static void i40e_restore_vlan(struct i40e_vsi *vsi)
 		i40e_vlan_stripping_disable(vsi);
 
 	for_each_set_bit(vid, vsi->active_vlans, VLAN_N_VID)
-		i40e_vlan_rx_add_vid(vsi->netdev, htons(ETH_P_8021Q),
-				     vid);
+		i40e_vlan_rx_add_vid_up(vsi->netdev, htons(ETH_P_8021Q),
+					vid);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 95a222d7ae43..a7b87f935411 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1925,7 +1925,8 @@ static void i40evf_reset_task(struct work_struct *work)
 	 * ndo_open() returning, so we can't assume it means all our open
 	 * tasks have finished, since we're not holding the rtnl_lock here.
 	 */
-	running = (adapter->state == __I40EVF_RUNNING);
+	running = ((adapter->state == __I40EVF_RUNNING) ||
+		   (adapter->state == __I40EVF_RESETTING));
 
 	if (running) {
 		netif_carrier_off(netdev);
-- 
2.17.0

Re: [net-next 00/11][pull request] 40GbE Intel Wired LAN Driver Updates 2018-05-14

[David Miller]

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Mon, 14 May 2018 08:27:36 -0700

> This series contains updates to virtchnl, i40e and i40evf.
> 
> Bruce cleans up whitespace and unnecessary parentheses in virtchnl.
> 
> Jake does a number of stat cleanups in the i40e driver, including
> cleanup of code indentation, whitespace issues, remove duplicate stats,
> fix grammar in code comment and general spring cleaning of the
> statistics code.
> 
> Patryk fixes an issue where we recalculate vectors left and vectors
> wanted but do not take into account the reduced number of queue pairs
> per VSI.
> 
> Harshitha adds tx_busy stat to ethtool stats to track the number of
> times we return NETDEV_TX_BUSY to the stack during transmit.
> 
> Paweł fixes a potential system crash when unloading the VF driver after
> a hardware reset.
> 
> The following are changes since commit 289e1f4e9e4a09c73a1c0152bb93855ea351ccda:
>   net: ipv4: ipconfig: fix unused variable
> and are available in the git repository at:
>   git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue 40GbE

Looks good, pulled, thanks Jeff.