[net-next 00/13][pull request] Intel Wired LAN Driver Updates 2014-06-09

[net-next 00/13][pull request] Intel Wired LAN Driver Updates 2014-06-09

[Jeff Kirsher]

This series contains more updates to i40e and i40evf.

Shannon adds checks for error status bits on the admin event queue and
provides notification if seen.  Cleans up unused variable and memory
allocation which was used earlier in driver development and is no longer
needed.  Also fixes the driver to not complain about removing
non-existent MAC addresses.  Bumps the driver versions for both i40e
and i40evf.

Catherine fixes a function header comment to make sure the comment correctly
reflects the function name.

Mitch adds code to allow for additional VSIs since the number of VSIs that
the firmware reports to us is a guaranteed minimum, not an absolute
maximum.  The hardware actually supports for more than the reported value,
which we often need.  Implements anti-spoofing for VFs for both MAC
addresses and VLANs, as well as enable this feature by default for all VFs.

Anjali changes the interrupt distribution policy to change the way
resources for special features are handled.  Fixes the driver to not fall
back to one queue if the only feature enabled is ATR, since FD_SB
and FD_ATR need to be checked independently in order to decide if we
will support multiple queue or not.  Allows the RSS table entry range
and GPS to be any number, not necessarily a power of 2 because hardware
does not restrict us to use a power of 2 GPS in the case of RSS as long as
we are not sharing the RSS table with another VSI (VMDq).

Frank modifies the driver to keep SR-IOV enabled in the case that RSS,
VMFq, FD_SB and DCB are disabled so that SR-IOV does not get turned off
unnecessarily.

Jesse fixes a bug in receive checksum where the driver was not marking
packets with bad checksums correctly, especially IPv6 packets with a bad
checksum.  To do this correctly, we need a define that may be set by
hardware in rare cases.

Greg fixes the driver to delete all the old and stale MAC filters for the
VF VSI when the host administrator changes the VF MAC address from under
its feet.

The following are changes since commit b78370c021c9d52721c7f96fbb3e10f5b2f428d3:
  Merge branch 'for-davem' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-next
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master

Anjali Singhai Jain (3):
  i40e: Changes to Interrupt distribution policy
  i40e: Do not fall back to one queue model if the only feature enabled
    is ATR
  i40e: Allow RSS table entry range and GPS to be any number, not
    necessarily power of 2

Catherine Sullivan (1):
  i40evf: Fix function header

Frank Zhang (1):
  i40e: keep SR-IOV enabled in the case that RSS, VMDQ, FD_SB and DCB
    are disabled

Greg Rose (1):
  i40e: Delete stale MAC filters after change

Jesse Brandeburg (1):
  i40e/i40evf: add PPRS bit to error bits and fix bug in Rx checksum

Mitch Williams (2):
  i40e: allow for more VSIs
  i40e: implement anti-spoofing for VFs

Shannon Nelson (4):
  i40e: add checks for AQ error status bits
  i40e: remove unused variable and memory allocation
  i40e: don't complain about removing non-existent addresses
  i40e/i40evf: bump version to 0.4.7 for i40e and 0.9.31 for i40evf

 drivers/net/ethernet/intel/i40e/i40e.h             |   3 +-
 drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c      |   2 +-
 drivers/net/ethernet/intel/i40e/i40e_debugfs.c     |   6 +-
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c     |   1 +
 drivers/net/ethernet/intel/i40e/i40e_main.c        | 177 ++++++++++++++-------
 drivers/net/ethernet/intel/i40e/i40e_txrx.c        |  76 ++++++---
 drivers/net/ethernet/intel/i40e/i40e_type.h        |   3 +-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c |  65 ++++++--
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h |   2 +
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c      |  74 ++++++---
 drivers/net/ethernet/intel/i40evf/i40e_type.h      |   3 +-
 drivers/net/ethernet/intel/i40evf/i40evf_main.c    |   4 +-
 12 files changed, 303 insertions(+), 113 deletions(-)

-- 
1.9.3

[net-next 01/13] i40e: add checks for AQ error status bits

[Jeff Kirsher]

From: Shannon Nelson <shannon.nelson@intel.com>

Check for error status bits on the AdminQ event queue and announce them
if seen.  If the Firmware sets these bits, it will trigger an AdminQ
interrupt to get the driver's attention to process the ARQ, which will
likely be enough to clear the actual issue.

Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Change-ID: I009e0ebc8be764e40e193b29aed2863f43eb5cb0
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 36 +++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 8c16e18..ab0b6e1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5132,11 +5132,47 @@ static void i40e_clean_adminq_subtask(struct i40e_pf *pf)
 	u16 pending, i = 0;
 	i40e_status ret;
 	u16 opcode;
+	u32 oldval;
 	u32 val;
 
 	if (!test_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state))
 		return;
 
+	/* check for error indications */
+	val = rd32(&pf->hw, pf->hw.aq.arq.len);
+	oldval = val;
+	if (val & I40E_PF_ARQLEN_ARQVFE_MASK) {
+		dev_info(&pf->pdev->dev, "ARQ VF Error detected\n");
+		val &= ~I40E_PF_ARQLEN_ARQVFE_MASK;
+	}
+	if (val & I40E_PF_ARQLEN_ARQOVFL_MASK) {
+		dev_info(&pf->pdev->dev, "ARQ Overflow Error detected\n");
+		val &= ~I40E_PF_ARQLEN_ARQOVFL_MASK;
+	}
+	if (val & I40E_PF_ARQLEN_ARQCRIT_MASK) {
+		dev_info(&pf->pdev->dev, "ARQ Critical Error detected\n");
+		val &= ~I40E_PF_ARQLEN_ARQCRIT_MASK;
+	}
+	if (oldval != val)
+		wr32(&pf->hw, pf->hw.aq.arq.len, val);
+
+	val = rd32(&pf->hw, pf->hw.aq.asq.len);
+	oldval = val;
+	if (val & I40E_PF_ATQLEN_ATQVFE_MASK) {
+		dev_info(&pf->pdev->dev, "ASQ VF Error detected\n");
+		val &= ~I40E_PF_ATQLEN_ATQVFE_MASK;
+	}
+	if (val & I40E_PF_ATQLEN_ATQOVFL_MASK) {
+		dev_info(&pf->pdev->dev, "ASQ Overflow Error detected\n");
+		val &= ~I40E_PF_ATQLEN_ATQOVFL_MASK;
+	}
+	if (val & I40E_PF_ATQLEN_ATQCRIT_MASK) {
+		dev_info(&pf->pdev->dev, "ASQ Critical Error detected\n");
+		val &= ~I40E_PF_ATQLEN_ATQCRIT_MASK;
+	}
+	if (oldval != val)
+		wr32(&pf->hw, pf->hw.aq.asq.len, val);
+
 	event.msg_size = I40E_MAX_AQ_BUF_SIZE;
 	event.msg_buf = kzalloc(event.msg_size, GFP_KERNEL);
 	if (!event.msg_buf)
-- 
1.9.3

[net-next 02/13] i40evf: Fix function header

[Jeff Kirsher]

From: Catherine Sullivan <catherine.sullivan@intel.com>

Fix function header comment to have the correct function name.

Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index e8d2481..af45d0c 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1395,7 +1395,7 @@ restart_watchdog:
 }
 
 /**
- * i40evf_configure_rss - increment to next available tx queue
+ * next_queue - increment to next available tx queue
  * @adapter: board private structure
  * @j: queue counter
  *
-- 
1.9.3

[net-next 03/13] i40e: allow for more VSIs

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

The number of VSIs that the firmware reports to us is a guaranteed
minimum, not an absolute maximum. The hardware actually supports far
more  than the reported value, which we often need.

To allow for this, we allocate space for a larger number of VSIs than is
guaranteed by the firmware, with the knowledge that we may fail to get
them all in the future.

Note that we are just allocating pointers here, the actual (much larger)
VSI structures are allocated on demand.

Change-ID: I6f4e535ce39d3bf417aef78306e04fbc7505140e
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h         |  2 +
 drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c  |  2 +-
 drivers/net/ethernet/intel/i40e/i40e_debugfs.c |  6 +--
 drivers/net/ethernet/intel/i40e/i40e_main.c    | 68 ++++++++++++++------------
 drivers/net/ethernet/intel/i40e/i40e_txrx.c    |  2 +-
 5 files changed, 45 insertions(+), 35 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index dc6d7c6..ebeaf95 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -72,6 +72,7 @@
 #define I40E_MIN_NUM_DESCRIPTORS      64
 #define I40E_MIN_MSIX                 2
 #define I40E_DEFAULT_NUM_VMDQ_VSI     8 /* max 256 VSIs */
+#define I40E_MIN_VSI_ALLOC            51 /* LAN, ATR, FCOE, 32 VF, 16 VMDQ */
 #define I40E_DEFAULT_QUEUES_PER_VMDQ  2 /* max 16 qps */
 #define I40E_DEFAULT_QUEUES_PER_VF    4
 #define I40E_DEFAULT_QUEUES_PER_TC    1 /* should be a power of 2 */
@@ -215,6 +216,7 @@ struct i40e_pf {
 	u16 rss_size;              /* num queues in the RSS array */
 	u16 rss_size_max;          /* HW defined max RSS queues */
 	u16 fdir_pf_filter_count;  /* num of guaranteed filters for this PF */
+	u16 num_alloc_vsi;         /* num VSIs this driver supports */
 	u8 atr_sample_rate;
 	bool wol_en;
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c b/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
index 6e8103a..871831a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_dcb_nl.c
@@ -232,7 +232,7 @@ static void i40e_dcbnl_del_app(struct i40e_pf *pf,
 			      struct i40e_ieee_app_priority_table *app)
 {
 	int v, err;
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v] && pf->vsi[v]->netdev) {
 			err = i40e_dcbnl_vsi_del_app(pf->vsi[v], app);
 			if (err)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
index 1bd0adb..cffdfc2 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_debugfs.c
@@ -45,7 +45,7 @@ static struct i40e_vsi *i40e_dbg_find_vsi(struct i40e_pf *pf, int seid)
 	if (seid < 0)
 		dev_info(&pf->pdev->dev, "%d: bad seid\n", seid);
 	else
-		for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+		for (i = 0; i < pf->num_alloc_vsi; i++)
 			if (pf->vsi[i] && (pf->vsi[i]->seid == seid))
 				return pf->vsi[i];
 
@@ -843,7 +843,7 @@ static void i40e_dbg_dump_vsi_no_seid(struct i40e_pf *pf)
 {
 	int i;
 
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i])
 			dev_info(&pf->pdev->dev, "dump vsi[%d]: %d\n",
 				 i, pf->vsi[i]->seid);
@@ -1526,7 +1526,7 @@ static ssize_t i40e_dbg_command_write(struct file *filp,
 			cnt = sscanf(&cmd_buf[15], "%i", &vsi_seid);
 			if (cnt == 0) {
 				int i;
-				for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+				for (i = 0; i < pf->num_alloc_vsi; i++)
 					i40e_vsi_reset_stats(pf->vsi[i]);
 				dev_info(&pf->pdev->dev, "vsi clear stats called for all vsi's\n");
 			} else if (cnt == 1) {
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index ab0b6e1..d9a6c69 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -652,7 +652,7 @@ static void i40e_update_link_xoff_rx(struct i40e_pf *pf)
 		return;
 
 	/* Clear the __I40E_HANG_CHECK_ARMED bit for all Tx rings */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 
 		if (!vsi || !vsi->tx_rings[0])
@@ -706,7 +706,7 @@ static void i40e_update_prio_xoff_rx(struct i40e_pf *pf)
 	}
 
 	/* Clear the __I40E_HANG_CHECK_ARMED bit for Tx rings */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 
 		if (!vsi || !vsi->tx_rings[0])
@@ -1734,7 +1734,7 @@ static void i40e_sync_filters_subtask(struct i40e_pf *pf)
 		return;
 	pf->flags &= ~I40E_FLAG_FILTER_SYNC;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v] &&
 		    (pf->vsi[v]->flags & I40E_VSI_FLAG_FILTER_CHANGED))
 			i40e_sync_vsi_filters(pf->vsi[v]);
@@ -3524,7 +3524,7 @@ static void i40e_clear_interrupt_scheme(struct i40e_pf *pf)
 	int i;
 
 	i40e_put_lump(pf->irq_pile, 0, I40E_PILE_VALID_BIT-1);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i])
 			i40e_vsi_free_q_vectors(pf->vsi[i]);
 	i40e_reset_interrupt_capability(pf);
@@ -3614,7 +3614,7 @@ static void i40e_pf_quiesce_all_vsi(struct i40e_pf *pf)
 {
 	int v;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			i40e_quiesce_vsi(pf->vsi[v]);
 	}
@@ -3628,7 +3628,7 @@ static void i40e_pf_unquiesce_all_vsi(struct i40e_pf *pf)
 {
 	int v;
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			i40e_unquiesce_vsi(pf->vsi[v]);
 	}
@@ -4069,7 +4069,7 @@ static void i40e_dcb_reconfigure(struct i40e_pf *pf)
 	}
 
 	/* Update each VSI */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (!pf->vsi[v])
 			continue;
 
@@ -4592,7 +4592,7 @@ void i40e_do_reset(struct i40e_pf *pf, u32 reset_flags)
 		/* Find the VSI(s) that requested a re-init */
 		dev_info(&pf->pdev->dev,
 			 "VSI reinit requested\n");
-		for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+		for (v = 0; v < pf->num_alloc_vsi; v++) {
 			struct i40e_vsi *vsi = pf->vsi[v];
 			if (vsi != NULL &&
 			    test_bit(__I40E_REINIT_REQUESTED, &vsi->state)) {
@@ -4919,7 +4919,7 @@ static void i40e_veb_link_event(struct i40e_veb *veb, bool link_up)
 			i40e_veb_link_event(pf->veb[i], link_up);
 
 	/* ... now the local VSIs */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && (pf->vsi[i]->uplink_seid == veb->seid))
 			i40e_vsi_link_event(pf->vsi[i], link_up);
 }
@@ -4976,7 +4976,7 @@ static void i40e_check_hang_subtask(struct i40e_pf *pf)
 	 *     for each q_vector
 	 *         force an interrupt
 	 */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		struct i40e_vsi *vsi = pf->vsi[v];
 		int armed = 0;
 
@@ -5026,7 +5026,7 @@ static void i40e_watchdog_subtask(struct i40e_pf *pf)
 	/* Update the stats for active netdevs so the network stack
 	 * can look at updated numbers whenever it cares to
 	 */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && pf->vsi[i]->netdev)
 			i40e_update_stats(pf->vsi[i]);
 
@@ -5278,7 +5278,7 @@ static int i40e_reconstitute_veb(struct i40e_veb *veb)
 	int ret;
 
 	/* build VSI that owns this VEB, temporarily attached to base VEB */
-	for (v = 0; v < pf->hw.func_caps.num_vsis && !ctl_vsi; v++) {
+	for (v = 0; v < pf->num_alloc_vsi && !ctl_vsi; v++) {
 		if (pf->vsi[v] &&
 		    pf->vsi[v]->veb_idx == veb->idx &&
 		    pf->vsi[v]->flags & I40E_VSI_FLAG_VEB_OWNER) {
@@ -5308,7 +5308,7 @@ static int i40e_reconstitute_veb(struct i40e_veb *veb)
 		goto end_reconstitute;
 
 	/* create the remaining VSIs attached to this VEB */
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (!pf->vsi[v] || pf->vsi[v] == ctl_vsi)
 			continue;
 
@@ -5421,7 +5421,7 @@ static void i40e_fdir_sb_setup(struct i40e_pf *pf)
 
 	/* find existing VSI and see if it needs configuring */
 	vsi = NULL;
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			vsi = pf->vsi[i];
 			break;
@@ -5451,7 +5451,7 @@ static void i40e_fdir_teardown(struct i40e_pf *pf)
 	int i;
 
 	i40e_fdir_filter_exit(pf);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			i40e_vsi_release(pf->vsi[i]);
 			break;
@@ -5480,7 +5480,7 @@ static int i40e_prep_for_reset(struct i40e_pf *pf)
 	/* quiesce the VSIs and their queues that are not already DOWN */
 	i40e_pf_quiesce_all_vsi(pf);
 
-	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
+	for (v = 0; v < pf->num_alloc_vsi; v++) {
 		if (pf->vsi[v])
 			pf->vsi[v]->seid = 0;
 	}
@@ -5960,15 +5960,15 @@ static int i40e_vsi_mem_alloc(struct i40e_pf *pf, enum i40e_vsi_type type)
 	 * find next empty vsi slot, looping back around if necessary
 	 */
 	i = pf->next_vsi;
-	while (i < pf->hw.func_caps.num_vsis && pf->vsi[i])
+	while (i < pf->num_alloc_vsi && pf->vsi[i])
 		i++;
-	if (i >= pf->hw.func_caps.num_vsis) {
+	if (i >= pf->num_alloc_vsi) {
 		i = 0;
 		while (i < pf->next_vsi && pf->vsi[i])
 			i++;
 	}
 
-	if (i < pf->hw.func_caps.num_vsis && !pf->vsi[i]) {
+	if (i < pf->num_alloc_vsi && !pf->vsi[i]) {
 		vsi_idx = i;             /* Found one! */
 	} else {
 		ret = -ENODEV;
@@ -7229,7 +7229,7 @@ int i40e_vsi_release(struct i40e_vsi *vsi)
 	 * the orphan VEBs yet.  We'll wait for an explicit remove request
 	 * from up the network stack.
 	 */
-	for (n = 0, i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (n = 0, i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] &&
 		    pf->vsi[i]->uplink_seid == uplink_seid &&
 		    (pf->vsi[i]->flags & I40E_VSI_FLAG_VEB_OWNER) == 0) {
@@ -7408,7 +7408,7 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type,
 
 	if (!veb && uplink_seid != pf->mac_seid) {
 
-		for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+		for (i = 0; i < pf->num_alloc_vsi; i++) {
 			if (pf->vsi[i] && pf->vsi[i]->seid == uplink_seid) {
 				vsi = pf->vsi[i];
 				break;
@@ -7651,7 +7651,7 @@ static void i40e_switch_branch_release(struct i40e_veb *branch)
 	 * NOTE: Removing the last VSI on a VEB has the SIDE EFFECT of removing
 	 *       the VEB itself, so don't use (*branch) after this loop.
 	 */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (!pf->vsi[i])
 			continue;
 		if (pf->vsi[i]->uplink_seid == branch_seid &&
@@ -7703,7 +7703,7 @@ void i40e_veb_release(struct i40e_veb *veb)
 	pf = veb->pf;
 
 	/* find the remaining VSI and check for extras */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->uplink_seid == veb->seid) {
 			n++;
 			vsi = pf->vsi[i];
@@ -7815,10 +7815,10 @@ struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags,
 	}
 
 	/* make sure there is such a vsi and uplink */
-	for (vsi_idx = 0; vsi_idx < pf->hw.func_caps.num_vsis; vsi_idx++)
+	for (vsi_idx = 0; vsi_idx < pf->num_alloc_vsi; vsi_idx++)
 		if (pf->vsi[vsi_idx] && pf->vsi[vsi_idx]->seid == vsi_seid)
 			break;
-	if (vsi_idx >= pf->hw.func_caps.num_vsis && vsi_seid != 0) {
+	if (vsi_idx >= pf->num_alloc_vsi && vsi_seid != 0) {
 		dev_info(&pf->pdev->dev, "vsi seid %d not found\n",
 			 vsi_seid);
 		return NULL;
@@ -8484,10 +8484,18 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	i40e_determine_queue_usage(pf);
 	i40e_init_interrupt_scheme(pf);
 
-	/* Set up the *vsi struct based on the number of VSIs in the HW,
-	 * and set up our local tracking of the MAIN PF vsi.
+	/* The number of VSIs reported by the FW is the minimum guaranteed
+	 * to us; HW supports far more and we share the remaining pool with
+	 * the other PFs. We allocate space for more than the guarantee with
+	 * the understanding that we might not get them all later.
 	 */
-	len = sizeof(struct i40e_vsi *) * pf->hw.func_caps.num_vsis;
+	if (pf->hw.func_caps.num_vsis < I40E_MIN_VSI_ALLOC)
+		pf->num_alloc_vsi = I40E_MIN_VSI_ALLOC;
+	else
+		pf->num_alloc_vsi = pf->hw.func_caps.num_vsis;
+
+	/* Set up the *vsi struct and our local tracking of the MAIN PF vsi. */
+	len = sizeof(struct i40e_vsi *) * pf->num_alloc_vsi;
 	pf->vsi = kzalloc(len, GFP_KERNEL);
 	if (!pf->vsi) {
 		err = -ENOMEM;
@@ -8500,7 +8508,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_vsis;
 	}
 	/* if FDIR VSI was set up, start it now */
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR) {
 			i40e_vsi_open(pf->vsi[i]);
 			break;
@@ -8695,7 +8703,7 @@ static void i40e_remove(struct pci_dev *pdev)
 
 	/* Clear all dynamic memory lists of rings, q_vectors, and VSIs */
 	i40e_clear_interrupt_scheme(pf);
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++) {
+	for (i = 0; i < pf->num_alloc_vsi; i++) {
 		if (pf->vsi[i]) {
 			i40e_vsi_clear_rings(pf->vsi[i]);
 			i40e_vsi_clear(pf->vsi[i]);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index d1a9a05..2622a86 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -62,7 +62,7 @@ int i40e_program_fdir_filter(struct i40e_fdir_filter *fdir_data, u8 *raw_packet,
 
 	/* find existing FDIR VSI */
 	vsi = NULL;
-	for (i = 0; i < pf->hw.func_caps.num_vsis; i++)
+	for (i = 0; i < pf->num_alloc_vsi; i++)
 		if (pf->vsi[i] && pf->vsi[i]->type == I40E_VSI_FDIR)
 			vsi = pf->vsi[i];
 	if (!vsi)
-- 
1.9.3

[net-next 04/13] i40e: remove unused variable and memory allocation

[Jeff Kirsher]

From: Shannon Nelson <shannon.nelson@intel.com>

This was a vestige of early driver development that no longer
has any actual use.

Change-ID: I95b5b19c4bbfaff8759197af671ebaf716cb6ab5
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h      |  1 -
 drivers/net/ethernet/intel/i40e/i40e_main.c | 10 ----------
 2 files changed, 11 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index ebeaf95..2af28fd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -297,7 +297,6 @@ struct i40e_pf {
 	u16 pf_seid;
 	u16 main_vsi_seid;
 	u16 mac_seid;
-	struct i40e_aqc_get_switch_config_data *sw_config;
 	struct kobject *switch_kobj;
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *i40e_dbg_pf;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index d9a6c69..4ba5192 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -7990,15 +7990,6 @@ int i40e_fetch_switch_configuration(struct i40e_pf *pf, bool printconfig)
 				 "header: %d reported %d total\n",
 				 num_reported, num_total);
 
-		if (num_reported) {
-			int sz = sizeof(*sw_config) * num_reported;
-
-			kfree(pf->sw_config);
-			pf->sw_config = kzalloc(sz, GFP_KERNEL);
-			if (pf->sw_config)
-				memcpy(pf->sw_config, sw_config, sz);
-		}
-
 		for (i = 0; i < num_reported; i++) {
 			struct i40e_aqc_switch_config_element_resp *ele =
 				&sw_config->element[i];
@@ -8718,7 +8709,6 @@ static void i40e_remove(struct pci_dev *pdev)
 
 	kfree(pf->qp_pile);
 	kfree(pf->irq_pile);
-	kfree(pf->sw_config);
 	kfree(pf->vsi);
 
 	/* force a PF reset to clean anything leftover */
-- 
1.9.3

[net-next 05/13] i40e: don't complain about removing non-existent addresses

[Jeff Kirsher]

From: Shannon Nelson <shannon.nelson@intel.com>

We don't need to complain in the log about mac addresses that
can't be deleted because they don't exist.

Change-ID: I4e6370df175bf72726f06d2206c03bcbfded8387
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 4ba5192..d0bc05f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1595,7 +1595,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 				num_del = 0;
 				memset(del_list, 0, sizeof(*del_list));
 
-				if (aq_ret)
+				if (aq_ret &&
+				    pf->hw.aq.asq_last_status !=
+							      I40E_AQ_RC_ENOENT)
 					dev_info(&pf->pdev->dev,
 						 "ignoring delete macvlan error, err %d, aq_err %d while flushing a full buffer\n",
 						 aq_ret,
@@ -1607,7 +1609,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 						     del_list, num_del, NULL);
 			num_del = 0;
 
-			if (aq_ret)
+			if (aq_ret &&
+			    pf->hw.aq.asq_last_status != I40E_AQ_RC_ENOENT)
 				dev_info(&pf->pdev->dev,
 					 "ignoring delete macvlan error, err %d, aq_err %d\n",
 					 aq_ret, pf->hw.aq.asq_last_status);
-- 
1.9.3

[net-next 06/13] i40e: implement anti-spoofing for VFs

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

Our hardware supports VF antispoofing for both MAC addresses and VLANs.
Enable this feature by default for all VFs and implement the netdev op
to control it from the command line.

Change-ID: Ifb941da22785848aa3aba6b2231be135b8ea8f31
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c        |  8 ++++
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 52 +++++++++++++++++++++-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h |  2 +
 3 files changed, 60 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index d0bc05f..afc37d3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -6887,6 +6887,7 @@ static const struct net_device_ops i40e_netdev_ops = {
 	.ndo_set_vf_rate	= i40e_ndo_set_vf_bw,
 	.ndo_get_vf_config	= i40e_ndo_get_vf_config,
 	.ndo_set_vf_link_state	= i40e_ndo_set_vf_link_state,
+	.ndo_set_vf_spoofchk	= i40e_ndo_set_vf_spoofck,
 #ifdef CONFIG_I40E_VXLAN
 	.ndo_add_vxlan_port	= i40e_add_vxlan_port,
 	.ndo_del_vxlan_port	= i40e_del_vxlan_port,
@@ -7121,6 +7122,13 @@ static int i40e_add_vsi(struct i40e_vsi *vsi)
 
 		ctxt.info.valid_sections |= cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
 		ctxt.info.port_vlan_flags |= I40E_AQ_VSI_PVLAN_MODE_ALL;
+		if (pf->vf[vsi->vf_id].spoofchk) {
+			ctxt.info.valid_sections |=
+				cpu_to_le16(I40E_AQ_VSI_PROP_SECURITY_VALID);
+			ctxt.info.sec_flags |=
+				(I40E_AQ_VSI_SEC_FLAG_ENABLE_VLAN_CHK |
+				 I40E_AQ_VSI_SEC_FLAG_ENABLE_MAC_CHK);
+		}
 		/* Setup the VSI tx/rx queue map for TC0 only for now */
 		i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, true);
 		break;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 4e7634c..3f6cad4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -899,6 +899,7 @@ int i40e_alloc_vfs(struct i40e_pf *pf, u16 num_alloc_vfs)
 		ret = -ENOMEM;
 		goto err_alloc;
 	}
+	pf->vf = vfs;
 
 	/* apply default profile */
 	for (i = 0; i < num_alloc_vfs; i++) {
@@ -908,13 +909,13 @@ int i40e_alloc_vfs(struct i40e_pf *pf, u16 num_alloc_vfs)
 
 		/* assign default capabilities */
 		set_bit(I40E_VIRTCHNL_VF_CAP_L2, &vfs[i].vf_caps);
+		vfs[i].spoofchk = true;
 		/* vf resources get allocated during reset */
 		i40e_reset_vf(&vfs[i], false);
 
 		/* enable vf vplan_qtable mappings */
 		i40e_enable_vf_mappings(&vfs[i]);
 	}
-	pf->vf = vfs;
 	pf->num_alloc_vfs = num_alloc_vfs;
 
 	i40e_enable_pf_switch_lb(pf);
@@ -2328,7 +2329,7 @@ int i40e_ndo_get_vf_config(struct net_device *netdev,
 		ivi->linkstate = IFLA_VF_LINK_STATE_ENABLE;
 	else
 		ivi->linkstate = IFLA_VF_LINK_STATE_DISABLE;
-
+	ivi->spoofchk = vf->spoofchk;
 	ret = 0;
 
 error_param:
@@ -2395,3 +2396,50 @@ int i40e_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link)
 error_out:
 	return ret;
 }
+
+/**
+ * i40e_ndo_set_vf_spoofchk
+ * @netdev: network interface device structure
+ * @vf_id: vf identifier
+ * @enable: flag to enable or disable feature
+ *
+ * Enable or disable VF spoof checking
+ **/
+int i40e_ndo_set_vf_spoofck(struct net_device *netdev, int vf_id, bool enable)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_vsi_context ctxt;
+	struct i40e_hw *hw = &pf->hw;
+	struct i40e_vf *vf;
+	int ret = 0;
+
+	/* validate the request */
+	if (vf_id >= pf->num_alloc_vfs) {
+		dev_err(&pf->pdev->dev, "Invalid VF Identifier %d\n", vf_id);
+		ret = -EINVAL;
+		goto out;
+	}
+
+	vf = &(pf->vf[vf_id]);
+
+	if (enable == vf->spoofchk)
+		goto out;
+
+	vf->spoofchk = enable;
+	memset(&ctxt, 0, sizeof(ctxt));
+	ctxt.seid = pf->vsi[vf->lan_vsi_index]->seid;
+	ctxt.pf_num = pf->hw.pf_id;
+	ctxt.info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_SECURITY_VALID);
+	if (enable)
+		ctxt.info.sec_flags |= I40E_AQ_VSI_SEC_FLAG_ENABLE_MAC_CHK;
+	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
+	if (ret) {
+		dev_err(&pf->pdev->dev, "Error %d updating VSI parameters\n",
+			ret);
+		ret = -EIO;
+	}
+out:
+	return ret;
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
index 5a559be..63e7e0d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
@@ -101,6 +101,7 @@ struct i40e_vf {
 	unsigned int tx_rate;	/* Tx bandwidth limit in Mbps */
 	bool link_forced;
 	bool link_up;		/* only valid if vf link is forced */
+	bool spoofchk;
 };
 
 void i40e_free_vfs(struct i40e_pf *pf);
@@ -121,6 +122,7 @@ int i40e_ndo_set_vf_bw(struct net_device *netdev, int vf_id, int min_tx_rate,
 int i40e_ndo_get_vf_config(struct net_device *netdev,
 			   int vf_id, struct ifla_vf_info *ivi);
 int i40e_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link);
+int i40e_ndo_set_vf_spoofck(struct net_device *netdev, int vf_id, bool enable);
 
 void i40e_vc_notify_link_state(struct i40e_pf *pf);
 void i40e_vc_notify_reset(struct i40e_pf *pf);
-- 
1.9.3

[net-next 07/13] i40e: Changes to Interrupt distribution policy

[Jeff Kirsher]

From: Anjali Singhai Jain <anjali.singhai@intel.com>

This patch changes the way resources are distributed to special features.

Change-ID: I847e49d714a1d70e97f3f994cb39bfb5e02ab016
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index afc37d3..eb32c5c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -6228,6 +6228,16 @@ static int i40e_init_msix(struct i40e_pf *pf)
 	for (i = 0; i < v_budget; i++)
 		pf->msix_entries[i].entry = i;
 	vec = i40e_reserve_msix_vectors(pf, v_budget);
+
+	if (vec != v_budget) {
+		/* If we have limited resources, we will start with no vectors
+		 * for the special features and then allocate vectors to some
+		 * of these features based on the policy and at the end disable
+		 * the features that did not get any vectors.
+		 */
+		pf->num_vmdq_msix = 0;
+	}
+
 	if (vec < I40E_MIN_MSIX) {
 		pf->flags &= ~I40E_FLAG_MSIX_ENABLED;
 		kfree(pf->msix_entries);
@@ -6236,27 +6246,25 @@ static int i40e_init_msix(struct i40e_pf *pf)
 
 	} else if (vec == I40E_MIN_MSIX) {
 		/* Adjust for minimal MSIX use */
-		dev_info(&pf->pdev->dev, "Features disabled, not enough MSI-X vectors\n");
-		pf->flags &= ~I40E_FLAG_VMDQ_ENABLED;
 		pf->num_vmdq_vsis = 0;
 		pf->num_vmdq_qps = 0;
-		pf->num_vmdq_msix = 0;
 		pf->num_lan_qps = 1;
 		pf->num_lan_msix = 1;
 
 	} else if (vec != v_budget) {
+		/* reserve the misc vector */
+		vec--;
+
 		/* Scale vector usage down */
 		pf->num_vmdq_msix = 1;    /* force VMDqs to only one vector */
-		vec--;                    /* reserve the misc vector */
+		pf->num_vmdq_vsis = 1;
 
 		/* partition out the remaining vectors */
 		switch (vec) {
 		case 2:
-			pf->num_vmdq_vsis = 1;
 			pf->num_lan_msix = 1;
 			break;
 		case 3:
-			pf->num_vmdq_vsis = 1;
 			pf->num_lan_msix = 2;
 			break;
 		default:
@@ -6268,6 +6276,11 @@ static int i40e_init_msix(struct i40e_pf *pf)
 		}
 	}
 
+	if ((pf->flags & I40E_FLAG_VMDQ_ENABLED) &&
+	    (pf->num_vmdq_msix == 0)) {
+		dev_info(&pf->pdev->dev, "VMDq disabled, not enough MSI-X vectors\n");
+		pf->flags &= ~I40E_FLAG_VMDQ_ENABLED;
+	}
 	return err;
 }
 
-- 
1.9.3

[net-next 08/13] i40e: keep SR-IOV enabled in the case that RSS, VMDQ, FD_SB and DCB are disabled

[Jeff Kirsher]

From: Frank Zhang <frank_1.zhang@intel.com>

Modify the logic in i40e_determine_queue_usage() so that
SR-IOV doesn't get turned off unnecessarily.

Change-ID: I86ca304fa9f742a50e9ea831b887f358a6a9d53d
Signed-off-by: Frank Zhang <frank_1.zhang@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index eb32c5c..b5f0dea 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8180,9 +8180,7 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf)
 	queues_left = pf->hw.func_caps.num_tx_qp;
 
 	if ((queues_left == 1) ||
-	    !(pf->flags & I40E_FLAG_MSIX_ENABLED) ||
-	    !(pf->flags & (I40E_FLAG_RSS_ENABLED | I40E_FLAG_FD_SB_ENABLED |
-			   I40E_FLAG_DCB_ENABLED))) {
+	    !(pf->flags & I40E_FLAG_MSIX_ENABLED)) {
 		/* one qp for PF, no queues for anything else */
 		queues_left = 0;
 		pf->rss_size = pf->num_lan_qps = 1;
@@ -8194,6 +8192,18 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf)
 			       I40E_FLAG_DCB_ENABLED	|
 			       I40E_FLAG_SRIOV_ENABLED	|
 			       I40E_FLAG_VMDQ_ENABLED);
+	} else if (!(pf->flags & (I40E_FLAG_RSS_ENABLED |
+				  I40E_FLAG_FD_SB_ENABLED |
+				  I40E_FLAG_DCB_ENABLED))) {
+		/* one qp for PF */
+		pf->rss_size = pf->num_lan_qps = 1;
+		queues_left -= pf->num_lan_qps;
+
+		pf->flags &= ~(I40E_FLAG_RSS_ENABLED	|
+			       I40E_FLAG_FD_SB_ENABLED	|
+			       I40E_FLAG_FD_ATR_ENABLED	|
+			       I40E_FLAG_DCB_ENABLED	|
+			       I40E_FLAG_VMDQ_ENABLED);
 	} else {
 		/* Not enough queues for all TCs */
 		if ((pf->flags & I40E_FLAG_DCB_ENABLED) &&
-- 
1.9.3

[net-next 09/13] i40e/i40evf: add PPRS bit to error bits and fix bug in Rx checksum

[Jeff Kirsher]

From: Jesse Brandeburg <jesse.brandeburg@intel.com>

The driver was not marking packets with bad checksums
correctly, especially IPv6 packets with a bad checksum.
To do this correctly we need a define that may be set by
hardware in rare cases.

Change-ID: I1a997b72b491ded27a78ac3bce1197b2d2611130
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c |  1 +
 drivers/net/ethernet/intel/i40e/i40e_txrx.c    | 74 +++++++++++++++++++-------
 drivers/net/ethernet/intel/i40e/i40e_type.h    |  3 +-
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c  | 74 +++++++++++++++++++-------
 drivers/net/ethernet/intel/i40evf/i40e_type.h  |  3 +-
 5 files changed, 115 insertions(+), 40 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index df3917b..b16c251 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -119,6 +119,7 @@ static struct i40e_stats i40e_gstrings_stats[] = {
 	I40E_PF_STAT("mac_local_faults", stats.mac_local_faults),
 	I40E_PF_STAT("mac_remote_faults", stats.mac_remote_faults),
 	I40E_PF_STAT("tx_timeout", tx_timeout_count),
+	I40E_PF_STAT("rx_csum_bad", hw_csum_rx_error),
 	I40E_PF_STAT("rx_length_errors", stats.rx_length_errors),
 	I40E_PF_STAT("link_xon_rx", stats.link_xon_rx),
 	I40E_PF_STAT("link_xoff_rx", stats.link_xoff_rx),
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 2622a86..d84f427 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1193,10 +1193,12 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				    u32 rx_error,
 				    u16 rx_ptype)
 {
+	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
+	bool ipv4 = false, ipv6 = false;
 	bool ipv4_tunnel, ipv6_tunnel;
 	__wsum rx_udp_csum;
-	__sum16 csum;
 	struct iphdr *iph;
+	__sum16 csum;
 
 	ipv4_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
 		      (rx_ptype < I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
@@ -1207,29 +1209,57 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	skb->ip_summed = CHECKSUM_NONE;
 
 	/* Rx csum enabled and ip headers found? */
-	if (!(vsi->netdev->features & NETIF_F_RXCSUM &&
-	      rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+	if (!(vsi->netdev->features & NETIF_F_RXCSUM))
+		return;
+
+	/* did the hardware decode the packet and checksum? */
+	if (!(rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+		return;
+
+	/* both known and outer_ip must be set for the below code to work */
+	if (!(decoded.known && decoded.outer_ip))
 		return;
 
+	if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+	    decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4)
+		ipv4 = true;
+	else if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+		 decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6)
+		ipv6 = true;
+
+	if (ipv4 &&
+	    (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
+			 (1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))))
+		goto checksum_fail;
+
 	/* likely incorrect csum if alternate IP extension headers found */
-	if (rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+	if (ipv6 &&
+	    decoded.inner_prot == I40E_RX_PTYPE_INNER_PROT_TCP &&
+	    rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT) &&
+	    rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+		/* don't increment checksum err here, non-fatal err */
 		return;
 
-	/* IP or L4 or outmost IP checksum error */
-	if (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_L4E_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))) {
-		vsi->back->hw_csum_rx_error++;
+	/* there was some L4 error, count error and punt packet to the stack */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT))
+		goto checksum_fail;
+
+	/* handle packets that were not able to be checksummed due
+	 * to arrival speed, in this case the stack can compute
+	 * the csum.
+	 */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_PPRS_SHIFT))
 		return;
-	}
 
+	/* If VXLAN traffic has an outer UDPv4 checksum we need to check
+	 * it in the driver, hardware does not do it for us.
+	 * Since L3L4P bit was set we assume a valid IHL value (>=5)
+	 * so the total length of IPv4 header is IHL*4 bytes
+	 * The UDP_0 bit *may* bet set if the *inner* header is UDP
+	 */
 	if (ipv4_tunnel &&
+	    (decoded.inner_prot != I40E_RX_PTYPE_INNER_PROT_UDP) &&
 	    !(rx_status & (1 << I40E_RX_DESC_STATUS_UDP_0_SHIFT))) {
-		/* If VXLAN traffic has an outer UDPv4 checksum we need to check
-		 * it in the driver, hardware does not do it for us.
-		 * Since L3L4P bit was set we assume a valid IHL value (>=5)
-		 * so the total length of IPv4 header is IHL*4 bytes
-		 */
 		skb->transport_header = skb->mac_header +
 					sizeof(struct ethhdr) +
 					(ip_hdr(skb)->ihl * 4);
@@ -1246,13 +1276,16 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				(skb->len - skb_transport_offset(skb)),
 				IPPROTO_UDP, rx_udp_csum);
 
-		if (udp_hdr(skb)->check != csum) {
-			vsi->back->hw_csum_rx_error++;
-			return;
-		}
+		if (udp_hdr(skb)->check != csum)
+			goto checksum_fail;
 	}
 
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	return;
+
+checksum_fail:
+	vsi->back->hw_csum_rx_error++;
 }
 
 /**
@@ -1429,6 +1462,9 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
 		/* ERR_MASK will only have valid bits if EOP set */
 		if (unlikely(rx_error & (1 << I40E_RX_DESC_ERROR_RXE_SHIFT))) {
 			dev_kfree_skb_any(skb);
+			/* TODO: shouldn't we increment a counter indicating the
+			 * drop?
+			 */
 			goto next_desc;
 		}
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index 6c977d2..42bfb2a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -541,7 +541,8 @@ enum i40e_rx_desc_error_bits {
 	I40E_RX_DESC_ERROR_IPE_SHIFT		= 3,
 	I40E_RX_DESC_ERROR_L4E_SHIFT		= 4,
 	I40E_RX_DESC_ERROR_EIPE_SHIFT		= 5,
-	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6
+	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6,
+	I40E_RX_DESC_ERROR_PPRS_SHIFT		= 7
 };
 
 enum i40e_rx_desc_error_l3l4e_fcoe_masks {
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index ae089df..48ebb6c 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -728,10 +728,12 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				    u32 rx_error,
 				    u16 rx_ptype)
 {
+	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(rx_ptype);
+	bool ipv4 = false, ipv6 = false;
 	bool ipv4_tunnel, ipv6_tunnel;
 	__wsum rx_udp_csum;
-	__sum16 csum;
 	struct iphdr *iph;
+	__sum16 csum;
 
 	ipv4_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
 		      (rx_ptype < I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
@@ -742,29 +744,57 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	skb->ip_summed = CHECKSUM_NONE;
 
 	/* Rx csum enabled and ip headers found? */
-	if (!(vsi->netdev->features & NETIF_F_RXCSUM &&
-	      rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+	if (!(vsi->netdev->features & NETIF_F_RXCSUM))
+		return;
+
+	/* did the hardware decode the packet and checksum? */
+	if (!(rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
+		return;
+
+	/* both known and outer_ip must be set for the below code to work */
+	if (!(decoded.known && decoded.outer_ip))
 		return;
 
+	if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+	    decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV4)
+		ipv4 = true;
+	else if (decoded.outer_ip == I40E_RX_PTYPE_OUTER_IP &&
+		 decoded.outer_ip_ver == I40E_RX_PTYPE_OUTER_IPV6)
+		ipv6 = true;
+
+	if (ipv4 &&
+	    (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
+			 (1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))))
+		goto checksum_fail;
+
 	/* likely incorrect csum if alternate IP extension headers found */
-	if (rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+	if (ipv6 &&
+	    decoded.inner_prot == I40E_RX_PTYPE_INNER_PROT_TCP &&
+	    rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT) &&
+	    rx_status & (1 << I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT))
+		/* don't increment checksum err here, non-fatal err */
 		return;
 
-	/* IP or L4 or outmost IP checksum error */
-	if (rx_error & ((1 << I40E_RX_DESC_ERROR_IPE_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_L4E_SHIFT) |
-			(1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))) {
-		vsi->back->hw_csum_rx_error++;
+	/* there was some L4 error, count error and punt packet to the stack */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT))
+		goto checksum_fail;
+
+	/* handle packets that were not able to be checksummed due
+	 * to arrival speed, in this case the stack can compute
+	 * the csum.
+	 */
+	if (rx_error & (1 << I40E_RX_DESC_ERROR_PPRS_SHIFT))
 		return;
-	}
 
+	/* If VXLAN traffic has an outer UDPv4 checksum we need to check
+	 * it in the driver, hardware does not do it for us.
+	 * Since L3L4P bit was set we assume a valid IHL value (>=5)
+	 * so the total length of IPv4 header is IHL*4 bytes
+	 * The UDP_0 bit *may* bet set if the *inner* header is UDP
+	 */
 	if (ipv4_tunnel &&
+	    (decoded.inner_prot != I40E_RX_PTYPE_INNER_PROT_UDP) &&
 	    !(rx_status & (1 << I40E_RX_DESC_STATUS_UDP_0_SHIFT))) {
-		/* If VXLAN traffic has an outer UDPv4 checksum we need to check
-		 * it in the driver, hardware does not do it for us.
-		 * Since L3L4P bit was set we assume a valid IHL value (>=5)
-		 * so the total length of IPv4 header is IHL*4 bytes
-		 */
 		skb->transport_header = skb->mac_header +
 					sizeof(struct ethhdr) +
 					(ip_hdr(skb)->ihl * 4);
@@ -781,13 +811,16 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				(skb->len - skb_transport_offset(skb)),
 				IPPROTO_UDP, rx_udp_csum);
 
-		if (udp_hdr(skb)->check != csum) {
-			vsi->back->hw_csum_rx_error++;
-			return;
-		}
+		if (udp_hdr(skb)->check != csum)
+			goto checksum_fail;
 	}
 
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	return;
+
+checksum_fail:
+	vsi->back->hw_csum_rx_error++;
 }
 
 /**
@@ -956,6 +989,9 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
 		/* ERR_MASK will only have valid bits if EOP set */
 		if (unlikely(rx_error & (1 << I40E_RX_DESC_ERROR_RXE_SHIFT))) {
 			dev_kfree_skb_any(skb);
+			/* TODO: shouldn't we increment a counter indicating the
+			 * drop?
+			 */
 			goto next_desc;
 		}
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_type.h b/drivers/net/ethernet/intel/i40evf/i40e_type.h
index e3c9ebb..0a7914d 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_type.h
@@ -541,7 +541,8 @@ enum i40e_rx_desc_error_bits {
 	I40E_RX_DESC_ERROR_IPE_SHIFT		= 3,
 	I40E_RX_DESC_ERROR_L4E_SHIFT		= 4,
 	I40E_RX_DESC_ERROR_EIPE_SHIFT		= 5,
-	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6
+	I40E_RX_DESC_ERROR_OVERSIZE_SHIFT	= 6,
+	I40E_RX_DESC_ERROR_PPRS_SHIFT		= 7
 };
 
 enum i40e_rx_desc_error_l3l4e_fcoe_masks {
-- 
1.9.3

[net-next 10/13] i40e: Do not fall back to one queue model if the only feature enabled is ATR

[Jeff Kirsher]

From: Anjali Singhai Jain <anjali.singhai@intel.com>

FD_SB and FD_ATR needs to be checked independently in order to decide if
we will support multiple queues or not.

Change-ID: I9d3274f5924c79e29efdbcf66a2fcca1fee2107f
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index b5f0dea..2f11542 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8194,6 +8194,7 @@ static void i40e_determine_queue_usage(struct i40e_pf *pf)
 			       I40E_FLAG_VMDQ_ENABLED);
 	} else if (!(pf->flags & (I40E_FLAG_RSS_ENABLED |
 				  I40E_FLAG_FD_SB_ENABLED |
+				  I40E_FLAG_FD_ATR_ENABLED |
 				  I40E_FLAG_DCB_ENABLED))) {
 		/* one qp for PF */
 		pf->rss_size = pf->num_lan_qps = 1;
-- 
1.9.3

[net-next 11/13] i40e: Delete stale MAC filters after change

[Jeff Kirsher]

From: Greg Rose <gregory.v.rose@intel.com>

Delete all the old and stale MAC filters for the VF VSI when the host
administrator changes the VF MAC address from under its feet.  Also don't
bother to add a filter for the VSI when its going to go away anyway.
Just record the new address and punch the VF reset.

Change-ID: Ic0d12055926f41989d1965ccf500053729c063ad
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 3f6cad4..385a46f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -2063,14 +2063,11 @@ int i40e_ndo_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
 	i40e_del_filter(vsi, vf->default_lan_addr.addr, vf->port_vlan_id,
 			true, false);
 
-	/* add the new mac address */
-	f = i40e_add_filter(vsi, mac, vf->port_vlan_id, true, false);
-	if (!f) {
-		dev_err(&pf->pdev->dev,
-			"Unable to add VF ucast filter\n");
-		ret = -ENOMEM;
-		goto error_param;
-	}
+	/* Delete all the filters for this VSI - we're going to kill it
+	 * anyway.
+	 */
+	list_for_each_entry(f, &vsi->mac_filter_list, list)
+		i40e_del_filter(vsi, f->macaddr, f->vlan, true, false);
 
 	dev_info(&pf->pdev->dev, "Setting MAC %pM on VF %d\n", mac, vf_id);
 	/* program mac filter */
-- 
1.9.3

[net-next 12/13] i40e: Allow RSS table entry range and GPS to be any number, not necessarily power of 2

[Jeff Kirsher]

From: Anjali Singhai Jain <anjali.singhai@intel.com>

We tell the HW upper boundary of power of 2 in VSI config,
but the HW does not restrict us to use just power of 2 GPS in
case of RSS as long as we are not sharing the RSS table with
another VSI (VMDq). We at present are not doing RSS in VMDq
VSI.
If we were to enable that and if the system had CPU count which
was not power 2, the VMDq VSIs will see a little skewed distribution.

Change-ID: I3ea797ce9065a3ca4fc4d04251bf195463410473
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 2f11542..adbeac4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1366,7 +1366,7 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi,
 	vsi->tc_config.numtc = numtc;
 	vsi->tc_config.enabled_tc = enabled_tc ? enabled_tc : 1;
 	/* Number of queues per enabled TC */
-	num_tc_qps = rounddown_pow_of_two(vsi->alloc_queue_pairs/numtc);
+	num_tc_qps = vsi->alloc_queue_pairs/numtc;
 	num_tc_qps = min_t(int, num_tc_qps, I40E_MAX_QUEUES_PER_TC);
 
 	/* Setup queue offset/count for all TCs for given VSI */
@@ -6498,7 +6498,6 @@ int i40e_reconfig_rss_queues(struct i40e_pf *pf, int queue_count)
 		return 0;
 
 	queue_count = min_t(int, queue_count, pf->rss_size_max);
-	queue_count = rounddown_pow_of_two(queue_count);
 
 	if (queue_count != pf->rss_size) {
 		i40e_prep_for_reset(pf);
@@ -6554,7 +6553,6 @@ static int i40e_sw_init(struct i40e_pf *pf)
 	if (pf->hw.func_caps.rss) {
 		pf->flags |= I40E_FLAG_RSS_ENABLED;
 		pf->rss_size = min_t(int, pf->rss_size_max, num_online_cpus());
-		pf->rss_size = rounddown_pow_of_two(pf->rss_size);
 	} else {
 		pf->rss_size = 1;
 	}
-- 
1.9.3

[net-next 13/13] i40e/i40evf: bump version to 0.4.7 for i40e and 0.9.31 for i40evf

[Jeff Kirsher]

From: Shannon Nelson <shannon.nelson@intel.com>

Bumpity and Fred Worm say it's time to change the numbers again.

Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Change-ID: I658731d022ea23cedede4be2bfecd8b4cc68d270
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c     | 2 +-
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index adbeac4..145cb9f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -39,7 +39,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 0
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 5
+#define DRV_VERSION_BUILD 7
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index af45d0c..d415785 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -36,7 +36,7 @@ char i40evf_driver_name[] = "i40evf";
 static const char i40evf_driver_string[] =
 	"Intel(R) XL710 X710 Virtual Function Network Driver";
 
-#define DRV_VERSION "0.9.29"
+#define DRV_VERSION "0.9.31"
 const char i40evf_driver_version[] = DRV_VERSION;
 static const char i40evf_copyright[] =
 	"Copyright (c) 2013 - 2014 Intel Corporation.";
-- 
1.9.3

Re: [net-next 01/13] i40e: add checks for AQ error status bits

[Sergei Shtylyov]

Hello.

On 06/09/2014 12:49 PM, Jeff Kirsher wrote:

> From: Shannon Nelson <shannon.nelson@intel.com>

> Check for error status bits on the AdminQ event queue and announce them
> if seen.  If the Firmware sets these bits, it will trigger an AdminQ
> interrupt to get the driver's attention to process the ARQ, which will
> likely be enough to clear the actual issue.

> Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
> Change-ID: I009e0ebc8be764e40e193b29aed2863f43eb5cb0
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> ---
>   drivers/net/ethernet/intel/i40e/i40e_main.c | 36 +++++++++++++++++++++++++++++
>   1 file changed, 36 insertions(+)

> diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
> index 8c16e18..ab0b6e1 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_main.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
> @@ -5132,11 +5132,47 @@ static void i40e_clean_adminq_subtask(struct i40e_pf *pf)
>   	u16 pending, i = 0;
>   	i40e_status ret;
>   	u16 opcode;
> +	u32 oldval;
>   	u32 val;
>
>   	if (!test_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state))
>   		return;
>
> +	/* check for error indications */
> +	val = rd32(&pf->hw, pf->hw.aq.arq.len);
> +	oldval = val;
> +	if (val & I40E_PF_ARQLEN_ARQVFE_MASK) {
> +		dev_info(&pf->pdev->dev, "ARQ VF Error detected\n");

    Hm, why not dev_err() here and below?

> +		val &= ~I40E_PF_ARQLEN_ARQVFE_MASK;
> +	}
> +	if (val & I40E_PF_ARQLEN_ARQOVFL_MASK) {
> +		dev_info(&pf->pdev->dev, "ARQ Overflow Error detected\n");
> +		val &= ~I40E_PF_ARQLEN_ARQOVFL_MASK;
> +	}
> +	if (val & I40E_PF_ARQLEN_ARQCRIT_MASK) {
> +		dev_info(&pf->pdev->dev, "ARQ Critical Error detected\n");
> +		val &= ~I40E_PF_ARQLEN_ARQCRIT_MASK;
> +	}
> +	if (oldval != val)
> +		wr32(&pf->hw, pf->hw.aq.arq.len, val);
> +
> +	val = rd32(&pf->hw, pf->hw.aq.asq.len);
> +	oldval = val;
> +	if (val & I40E_PF_ATQLEN_ATQVFE_MASK) {
> +		dev_info(&pf->pdev->dev, "ASQ VF Error detected\n");
> +		val &= ~I40E_PF_ATQLEN_ATQVFE_MASK;
> +	}
> +	if (val & I40E_PF_ATQLEN_ATQOVFL_MASK) {
> +		dev_info(&pf->pdev->dev, "ASQ Overflow Error detected\n");
> +		val &= ~I40E_PF_ATQLEN_ATQOVFL_MASK;
> +	}
> +	if (val & I40E_PF_ATQLEN_ATQCRIT_MASK) {
> +		dev_info(&pf->pdev->dev, "ASQ Critical Error detected\n");
> +		val &= ~I40E_PF_ATQLEN_ATQCRIT_MASK;
> +	}
> +	if (oldval != val)
> +		wr32(&pf->hw, pf->hw.aq.asq.len, val);
> +

WBR, Sergei

Re: [net-next 06/13] i40e: implement anti-spoofing for VFs

[Or Gerlitz]

On Mon, Jun 9, 2014 at 11:49 AM, Jeff Kirsher
<jeffrey.t.kirsher@intel.com> wrote:
> From: Mitch Williams <mitch.a.williams@intel.com>
>
> Our hardware supports VF antispoofing for both MAC addresses and VLANs.
> Enable this feature by default for all VFs

What do you expect the HW to do when spoof check is enabled (by
default) but the admin didn't configure a MAC address for the VF
through the PF? that is the VF is allowed to use what ever MAC they
want to?

> and implement the netdev op to control it from the command line.

Re: [net-next 01/13] i40e: add checks for AQ error status bits

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 2977 bytes --]

On Mon, 2014-06-09 at 17:21 +0400, Sergei Shtylyov wrote:
> Hello.
> 
> On 06/09/2014 12:49 PM, Jeff Kirsher wrote:
> 
> > From: Shannon Nelson <shannon.nelson@intel.com>
> 
> > Check for error status bits on the AdminQ event queue and announce them
> > if seen.  If the Firmware sets these bits, it will trigger an AdminQ
> > interrupt to get the driver's attention to process the ARQ, which will
> > likely be enough to clear the actual issue.
> 
> > Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
> > Change-ID: I009e0ebc8be764e40e193b29aed2863f43eb5cb0
> > Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> > ---
> >   drivers/net/ethernet/intel/i40e/i40e_main.c | 36 +++++++++++++++++++++++++++++
> >   1 file changed, 36 insertions(+)
> 
> > diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
> > index 8c16e18..ab0b6e1 100644
> > --- a/drivers/net/ethernet/intel/i40e/i40e_main.c
> > +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
> > @@ -5132,11 +5132,47 @@ static void i40e_clean_adminq_subtask(struct i40e_pf *pf)
> >   	u16 pending, i = 0;
> >   	i40e_status ret;
> >   	u16 opcode;
> > +	u32 oldval;
> >   	u32 val;
> >
> >   	if (!test_bit(__I40E_ADMINQ_EVENT_PENDING, &pf->state))
> >   		return;
> >
> > +	/* check for error indications */
> > +	val = rd32(&pf->hw, pf->hw.aq.arq.len);
> > +	oldval = val;
> > +	if (val & I40E_PF_ARQLEN_ARQVFE_MASK) {
> > +		dev_info(&pf->pdev->dev, "ARQ VF Error detected\n");
> 
>     Hm, why not dev_err() here and below?

The thought was that these should be more of "FYI..." type of messages
not "Oh Crap!..." messages.  So that is why dev_err() was not used,
although we are not opposed to changing it if you feel it warrants it in
a follow-up patch.

> 
> > +		val &= ~I40E_PF_ARQLEN_ARQVFE_MASK;
> > +	}
> > +	if (val & I40E_PF_ARQLEN_ARQOVFL_MASK) {
> > +		dev_info(&pf->pdev->dev, "ARQ Overflow Error detected\n");
> > +		val &= ~I40E_PF_ARQLEN_ARQOVFL_MASK;
> > +	}
> > +	if (val & I40E_PF_ARQLEN_ARQCRIT_MASK) {
> > +		dev_info(&pf->pdev->dev, "ARQ Critical Error detected\n");
> > +		val &= ~I40E_PF_ARQLEN_ARQCRIT_MASK;
> > +	}
> > +	if (oldval != val)
> > +		wr32(&pf->hw, pf->hw.aq.arq.len, val);
> > +
> > +	val = rd32(&pf->hw, pf->hw.aq.asq.len);
> > +	oldval = val;
> > +	if (val & I40E_PF_ATQLEN_ATQVFE_MASK) {
> > +		dev_info(&pf->pdev->dev, "ASQ VF Error detected\n");
> > +		val &= ~I40E_PF_ATQLEN_ATQVFE_MASK;
> > +	}
> > +	if (val & I40E_PF_ATQLEN_ATQOVFL_MASK) {
> > +		dev_info(&pf->pdev->dev, "ASQ Overflow Error detected\n");
> > +		val &= ~I40E_PF_ATQLEN_ATQOVFL_MASK;
> > +	}
> > +	if (val & I40E_PF_ATQLEN_ATQCRIT_MASK) {
> > +		dev_info(&pf->pdev->dev, "ASQ Critical Error detected\n");
> > +		val &= ~I40E_PF_ATQLEN_ATQCRIT_MASK;
> > +	}
> > +	if (oldval != val)
> > +		wr32(&pf->hw, pf->hw.aq.asq.len, val);
> > +
> 
> WBR, Sergei
> 



[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [net-next 01/13] i40e: add checks for AQ error status bits

[Joe Perches]

On Mon, 2014-06-09 at 13:35 -0700, Jeff Kirsher wrote:
> On Mon, 2014-06-09 at 17:21 +0400, Sergei Shtylyov wrote:
> > On 06/09/2014 12:49 PM, Jeff Kirsher wrote:
> > > From: Shannon Nelson <shannon.nelson@intel.com>

> > > If the Firmware sets these bits, it will trigger an AdminQ
> > > interrupt to get the driver's attention to process the ARQ, which will
> > > likely be enough to clear the actual issue.
 
> >     Hm, why not dev_err() here and below?

> The thought was that these should be more of "FYI..." type of messages
> not "Oh Crap!..." messages.  So that is why dev_err() was not used,
> although we are not opposed to changing it if you feel it warrants it in
> a follow-up patch.
[]
> > > +	if (val & I40E_PF_ATQLEN_ATQCRIT_MASK) {
> > > +		dev_info(&pf->pdev->dev, "ASQ Critical Error detected\n");
> > > +		val &= ~I40E_PF_ATQLEN_ATQCRIT_MASK;
> > > +	}

I thought it was odd to have a "critical error"
emitted at KERN_INFO

Maybe adding something like
	"ARQ should fix this automatically"
would be enough.

Should any/all of these be ratelimited or maybe even
changed to dev_dbg?

Re: [net-next 01/13] i40e: add checks for AQ error status bits

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 1675 bytes --]

On Mon, 2014-06-09 at 14:02 -0700, Joe Perches wrote:
> On Mon, 2014-06-09 at 13:35 -0700, Jeff Kirsher wrote:
> > On Mon, 2014-06-09 at 17:21 +0400, Sergei Shtylyov wrote:
> > > On 06/09/2014 12:49 PM, Jeff Kirsher wrote:
> > > > From: Shannon Nelson <shannon.nelson@intel.com>
> 
> > > > If the Firmware sets these bits, it will trigger an AdminQ
> > > > interrupt to get the driver's attention to process the ARQ, which will
> > > > likely be enough to clear the actual issue.
>  
> > >     Hm, why not dev_err() here and below?
> 
> > The thought was that these should be more of "FYI..." type of messages
> > not "Oh Crap!..." messages.  So that is why dev_err() was not used,
> > although we are not opposed to changing it if you feel it warrants it in
> > a follow-up patch.
> []
> > > > +	if (val & I40E_PF_ATQLEN_ATQCRIT_MASK) {
> > > > +		dev_info(&pf->pdev->dev, "ASQ Critical Error detected\n");
> > > > +		val &= ~I40E_PF_ATQLEN_ATQCRIT_MASK;
> > > > +	}
> 
> I thought it was odd to have a "critical error"
> emitted at KERN_INFO
> 
> Maybe adding something like
> 	"ARQ should fix this automatically"
> would be enough.
> 
> Should any/all of these be ratelimited or maybe even
> changed to dev_dbg?
> 
> 

Well, as Shannon noted in the patch description:
"If the Firmware sets these bits, it will trigger an AdminQ
interrupt to get the driver's attention to process the ARQ, which will
likely be enough to clear the actual issue."

To me that means the log messages should not be filling up with these
types of messages and if they do occur, it should be transient and
resolve itself.  So does that warrant dev_dbg? 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [net-next 01/13] i40e: add checks for AQ error status bits

[Joe Perches]

On Mon, 2014-06-09 at 14:10 -0700, Jeff Kirsher wrote:
> Well, as Shannon noted in the patch description:
> "If the Firmware sets these bits, it will trigger an AdminQ
> interrupt to get the driver's attention to process the ARQ, which will
> likely be enough to clear the actual issue."

Yeah, I quoted that too.

> To me that means the log messages should not be filling up with these
> types of messages and if they do occur, it should be transient and
> resolve itself.  So does that warrant dev_dbg? 

No idea.  That's up to you.  They're your messages.

I guess it depends on what you expect your users to
do when these are emitted.

Re: [net-next 00/13][pull request] Intel Wired LAN Driver Updates 2014-06-09

[David Miller]

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Mon,  9 Jun 2014 01:49:05 -0700

> This series contains more updates to i40e and i40evf.

Pulled, thanks Jeff.

Please make sure the feedback given for patches #1 and #6 get addressed.

Re: [net-next 00/13][pull request] Intel Wired LAN Driver Updates 2014-06-09

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 529 bytes --]

On Tue, 2014-06-10 at 20:26 -0700, David Miller wrote:
> From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> Date: Mon,  9 Jun 2014 01:49:05 -0700
> 
> > This series contains more updates to i40e and i40evf.
> 
> Pulled, thanks Jeff.
> 
> Please make sure the feedback given for patches #1 and #6 get
> addressed.

Thanks Dave, since Mitch is on vacation, I have someone looking to
answer Or's questions regarding patch 6.  Shannon and I are looking at
the feedback from patch 1 to see if changes are needed or not.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [net-next 00/13][pull request] Intel Wired LAN Driver Updates 2014-06-09

[Or Gerlitz]

On Wed, Jun 11, 2014 at 1:16 PM, Jeff Kirsher
<jeffrey.t.kirsher@intel.com> wrote:
> On Tue, 2014-06-10 at 20:26 -0700, David Miller wrote:
>> From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
>> Date: Mon,  9 Jun 2014 01:49:05 -0700
>>
>> > This series contains more updates to i40e and i40evf.
>>
>> Pulled, thanks Jeff.
>>
>> Please make sure the feedback given for patches #1 and #6 get
>> addressed.
>
> Thanks Dave, since Mitch is on vacation, I have someone looking to
> answer Or's questions regarding patch 6.  Shannon and I are looking at
> the feedback from patch 1 to see if changes are needed or not.

Jeef, please note that also this feedback which I gave on earlier
patch remained unanswered
http://marc.info/?l=linux-netdev&m=140117707628190&w=2

Re: [net-next 00/13][pull request] Intel Wired LAN Driver Updates 2014-06-09

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 949 bytes --]

On Wed, 2014-06-11 at 13:42 +0300, Or Gerlitz wrote:
> On Wed, Jun 11, 2014 at 1:16 PM, Jeff Kirsher
> <jeffrey.t.kirsher@intel.com> wrote:
> > On Tue, 2014-06-10 at 20:26 -0700, David Miller wrote:
> >> From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> >> Date: Mon,  9 Jun 2014 01:49:05 -0700
> >>
> >> > This series contains more updates to i40e and i40evf.
> >>
> >> Pulled, thanks Jeff.
> >>
> >> Please make sure the feedback given for patches #1 and #6 get
> >> addressed.
> >
> > Thanks Dave, since Mitch is on vacation, I have someone looking to
> > answer Or's questions regarding patch 6.  Shannon and I are looking at
> > the feedback from patch 1 to see if changes are needed or not.
> 
> Jeef, please note that also this feedback which I gave on earlier
> patch remained unanswered
> http://marc.info/?l=linux-netdev&m=140117707628190&w=2

Oh yeah, thanks for the reminder Or.  I will ping Jesse on that one.


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [net-next 06/13] i40e: implement anti-spoofing for VFs

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 1087 bytes --]

On Mon, 2014-06-09 at 22:49 +0300, Or Gerlitz wrote:
> On Mon, Jun 9, 2014 at 11:49 AM, Jeff Kirsher
> <jeffrey.t.kirsher@intel.com> wrote:
> > From: Mitch Williams <mitch.a.williams@intel.com>
> >
> > Our hardware supports VF antispoofing for both MAC addresses and VLANs.
> > Enable this feature by default for all VFs
> 
> What do you expect the HW to do when spoof check is enabled (by
> default) but the admin didn't configure a MAC address for the VF
> through the PF? that is the VF is allowed to use what ever MAC they
> want to?
> 
> > and implement the netdev op to control it from the command line.

Here is the answer I got:
If the VF mac address is set within the VM and it is accepted by the PF,
than any packets with that mac address would be allowed out of the
interface.

If the VF attempts to send a packet with a mac address that has not been
sent to and accepted/configured by the PF than this would get blocked by
the anti-spoof detection.

The VF mac address must be configured by the PF in either case (set in
the host or set in the VM).

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [net-next 06/13] i40e: implement anti-spoofing for VFs

[Or Gerlitz]

On Wed, Jun 11, 2014 at 3:13 PM, Jeff Kirsher
<jeffrey.t.kirsher@intel.com> wrote:
> On Mon, 2014-06-09 at 22:49 +0300, Or Gerlitz wrote:
>> On Mon, Jun 9, 2014 at 11:49 AM, Jeff Kirsher
>> <jeffrey.t.kirsher@intel.com> wrote:
>> > From: Mitch Williams <mitch.a.williams@intel.com>
>> >
>> > Our hardware supports VF antispoofing for both MAC addresses and VLANs.
>> > Enable this feature by default for all VFs
>>
>> What do you expect the HW to do when spoof check is enabled (by
>> default) but the admin didn't configure a MAC address for the VF
>> through the PF? that is the VF is allowed to use what ever MAC they
>> want to?
>>
>> > and implement the netdev op to control it from the command line.
>
> Here is the answer I got:
> If the VF mac address is set within the VM and it is accepted by the PF,

When the admin doesn't configure MAC address for the VM, what logic is
applied by the PF to decide whether or not to accept a VF MAC set by the VM?

Re: [net-next 06/13] i40e: implement anti-spoofing for VFs

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 1319 bytes --]

On Wed, 2014-06-11 at 15:43 +0300, Or Gerlitz wrote:
> On Wed, Jun 11, 2014 at 3:13 PM, Jeff Kirsher
> <jeffrey.t.kirsher@intel.com> wrote:
> > On Mon, 2014-06-09 at 22:49 +0300, Or Gerlitz wrote:
> >> On Mon, Jun 9, 2014 at 11:49 AM, Jeff Kirsher
> >> <jeffrey.t.kirsher@intel.com> wrote:
> >> > From: Mitch Williams <mitch.a.williams@intel.com>
> >> >
> >> > Our hardware supports VF antispoofing for both MAC addresses and VLANs.
> >> > Enable this feature by default for all VFs
> >>
> >> What do you expect the HW to do when spoof check is enabled (by
> >> default) but the admin didn't configure a MAC address for the VF
> >> through the PF? that is the VF is allowed to use what ever MAC they
> >> want to?
> >>
> >> > and implement the netdev op to control it from the command line.
> >
> > Here is the answer I got:
> > If the VF mac address is set within the VM and it is accepted by the PF,
> 
> When the admin doesn't configure MAC address for the VM, what logic is
> applied by the PF to decide whether or not to accept a VF MAC set by the VM?

If the PF has not set the MAC address, it will check for a valid MAC
address that is neither broadcast or all zeros.

Look at the add_addr message handler i40e_vc_add_mac_addr_msg() or more
specifically i40e_check_vf_permission(). 

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]