[net-next v3 00/16][pull request] Intel Wired LAN Driver Updates

[net-next v3 00/16][pull request] Intel Wired LAN Driver Updates

[Jeff Kirsher]

This series contains updates to i40e only.

Anjali provides a patch to prevent messages from stray HMC events, except
at interrupt message level, and refactors the HMC error handling.
  
Catherine adds routines in probe to populate/check PCI bus speed and width,
then verify we are in a 8GT/s x8 PCIe slot and warn when we are not.
  
Shannon adds Wake-on-LAN support for i40e, fixes curly brace use as well as
return type for i40e_vsi_clear_rings().
  
Joseph implements receive offload for VXLAN for i40e, where the hardware
supports checksum offload/verification of the inner/outer header.

Mitch provides the bulk of the changes, where he refactors the VF reset
code so that it works on real hardware.  Then does code cleanup by
calling existing functions to enable and disable queues for VFs and
remove unused functions.  Removes a unnecessary log messages that are
seen at every VF reset, for example complaining about disabling queues
that are already disabled.  Fixes an error return when the VF asks to
add an invalid MAC address and if the VF sends a bad message, make it
more informative about what is actually going on.

Jesse refactors the LED function to flash LED lights correctly.

v2:
 - removed patch 5 "i40e: add set settings and pauseparam" based on
   feedback from Ben Hutchings, will re-work that patch for later
   submission
 - Added patch "i40e: Implementation of vxlan ndo's" from Joseph to
   address Or Gerlitz's questions and concerns.  This patch adds the
   implementation for the VXLAN ndo's and allows the hardware to do
   receive checksum offload for inner packets on the UDP ports that
   VXLAN notifies us about.
 - Added patch "i40e: using for_each_set_bit to simplify the code"
   from Wei Yongjun.  This patch uses for_each_set_bit() to simply
   the code.

v3:
 - fixed indentation issue in patch 11 based on feedback from
   Sergei Shtylyov.

The following are changes since commit 852ad5e631967ae2203cb08c5b6b42c26011ed63:
  Merge branch 'bridge_cleanups'
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master

Anjali Singhai Jain (1):
  i40e: Suppress HMC error to Interrupt message level

Catherine Sullivan (1):
  i40e: Populate and check pci bus speed and width

Jesse Brandeburg (1):
  i40e: update led set args

Joseph Gasparakis (2):
  i40e: Implementation of vxlan ndo's
  i40e: Rx checksum offload for VXLAN

Mitch Williams (8):
  i40e: move i40e_reset_vf
  i40e: refactor VF reset flow
  i40e: remove redundant code
  i40e: remove chatty log messages
  i40e: fix error return
  i40e: be more informative
  i40e: make a define from a large constant
  i40e: report VF MAC addresses correctly

Shannon Nelson (2):
  i40e: add wake-on-lan support
  i40e: fix curly brace use and return type

Wei Yongjun (1):
  i40e: using for_each_set_bit to simplify the code

 drivers/net/ethernet/intel/Kconfig                 |   1 +
 drivers/net/ethernet/intel/i40e/i40e.h             |   5 +
 drivers/net/ethernet/intel/i40e/i40e_common.c      | 193 ++++++-
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c     |  48 +-
 drivers/net/ethernet/intel/i40e/i40e_main.c        | 195 +++++--
 drivers/net/ethernet/intel/i40e/i40e_prototype.h   |   9 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.c        |  57 +-
 drivers/net/ethernet/intel/i40e/i40e_txrx.h        |  20 +
 drivers/net/ethernet/intel/i40e/i40e_type.h        |  62 ++-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 581 ++++++---------------
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h |   2 +-
 11 files changed, 635 insertions(+), 538 deletions(-)

-- 
1.8.3.1

[net-next v3 01/16] i40e: using for_each_set_bit to simplify the code

[Jeff Kirsher]

From: Wei Yongjun <yongjun_wei@trendmicro.com.cn>

Using for_each_set_bit() to simplify the code.

Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 64 ++++------------------
 1 file changed, 12 insertions(+), 52 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 68e1f8e..1ea0886 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -260,23 +260,17 @@ static void i40e_config_irq_link_list(struct i40e_vf *vf, u16 vsi_idx,
 		goto irq_list_done;
 	}
 	tempmap = vecmap->rxq_map;
-	vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (vsi_queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(vsi_queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		linklistmap |= (1 <<
 				(I40E_VIRTCHNL_SUPPORTED_QTYPES *
 				 vsi_queue_id));
-		vsi_queue_id =
-		    find_next_bit(&tempmap, I40E_MAX_VSI_QP, vsi_queue_id + 1);
 	}
 
 	tempmap = vecmap->txq_map;
-	vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (vsi_queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(vsi_queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		linklistmap |= (1 <<
 				(I40E_VIRTCHNL_SUPPORTED_QTYPES * vsi_queue_id
 				 + 1));
-		vsi_queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					     vsi_queue_id + 1);
 	}
 
 	next_q = find_first_bit(&linklistmap,
@@ -1293,27 +1287,21 @@ static int i40e_vc_config_irq_map_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 
 		/* lookout for the invalid queue index */
 		tempmap = map->rxq_map;
-		vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-		while (vsi_queue_id < I40E_MAX_VSI_QP) {
+		for_each_set_bit(vsi_queue_id, &tempmap, I40E_MAX_VSI_QP) {
 			if (!i40e_vc_isvalid_queue_id(vf, vsi_id,
 						      vsi_queue_id)) {
 				aq_ret = I40E_ERR_PARAM;
 				goto error_param;
 			}
-			vsi_queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-						     vsi_queue_id + 1);
 		}
 
 		tempmap = map->txq_map;
-		vsi_queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-		while (vsi_queue_id < I40E_MAX_VSI_QP) {
+		for_each_set_bit(vsi_queue_id, &tempmap, I40E_MAX_VSI_QP) {
 			if (!i40e_vc_isvalid_queue_id(vf, vsi_id,
 						      vsi_queue_id)) {
 				aq_ret = I40E_ERR_PARAM;
 				goto error_param;
 			}
-			vsi_queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-						     vsi_queue_id + 1);
 		}
 
 		i40e_config_irq_link_list(vf, vsi_id, map);
@@ -1358,31 +1346,23 @@ static int i40e_vc_enable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	}
 
 	tempmap = vqs->rx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
 			aq_ret = I40E_ERR_PARAM;
 			goto error_param;
 		}
 		i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
 				       I40E_QUEUE_CTRL_ENABLE);
-
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 	tempmap = vqs->tx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
 			aq_ret = I40E_ERR_PARAM;
 			goto error_param;
 		}
 		i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
 				       I40E_QUEUE_CTRL_ENABLE);
-
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 	/* Poll the status register to make sure that the
@@ -1391,29 +1371,23 @@ static int i40e_vc_enable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	udelay(10);
 
 	tempmap = vqs->rx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
 					   I40E_QUEUE_CTRL_ENABLECHECK)) {
 			dev_err(&pf->pdev->dev,
 				"Queue control check failed on RX queue %d of VSI %d VF %d\n",
 				queue_id, vsi_id, vf->vf_id);
 		}
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 	tempmap = vqs->tx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
 					   I40E_QUEUE_CTRL_ENABLECHECK)) {
 			dev_err(&pf->pdev->dev,
 				"Queue control check failed on TX queue %d of VSI %d VF %d\n",
 				queue_id, vsi_id, vf->vf_id);
 		}
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 error_param:
@@ -1457,31 +1431,23 @@ static int i40e_vc_disable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	}
 
 	tempmap = vqs->rx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
 			aq_ret = I40E_ERR_PARAM;
 			goto error_param;
 		}
 		i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
 				       I40E_QUEUE_CTRL_DISABLE);
-
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 	tempmap = vqs->tx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
 			aq_ret = I40E_ERR_PARAM;
 			goto error_param;
 		}
 		i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
 				       I40E_QUEUE_CTRL_DISABLE);
-
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 	/* Poll the status register to make sure that the
@@ -1490,29 +1456,23 @@ static int i40e_vc_disable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	udelay(10);
 
 	tempmap = vqs->rx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
 					   I40E_QUEUE_CTRL_DISABLECHECK)) {
 			dev_err(&pf->pdev->dev,
 				"Queue control check failed on RX queue %d of VSI %d VF %d\n",
 				queue_id, vsi_id, vf->vf_id);
 		}
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 	tempmap = vqs->tx_queues;
-	queue_id = find_first_bit(&tempmap, I40E_MAX_VSI_QP);
-	while (queue_id < I40E_MAX_VSI_QP) {
+	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
 		if (i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
 					   I40E_QUEUE_CTRL_DISABLECHECK)) {
 			dev_err(&pf->pdev->dev,
 				"Queue control check failed on TX queue %d of VSI %d VF %d\n",
 				queue_id, vsi_id, vf->vf_id);
 		}
-		queue_id = find_next_bit(&tempmap, I40E_MAX_VSI_QP,
-					 queue_id + 1);
 	}
 
 error_param:
-- 
1.8.3.1

[net-next v3 02/16] i40e: Suppress HMC error to Interrupt message level

[Jeff Kirsher]

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

Prevent messages from stray HMC events, except at interrupt message
level, and refactor HMC error handling.

Change-Id: Iabd203ba1dfc26a136b638597f3e9991acfa29f3
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index efdf8a2..f14b31c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2814,6 +2814,11 @@ static irqreturn_t i40e_intr(int irq, void *data)
 			pf->empr_count++;
 	}
 
+	if (icr0 & I40E_PFINT_ICR0_HMC_ERR_MASK) {
+		icr0 &= ~I40E_PFINT_ICR0_HMC_ERR_MASK;
+		dev_info(&pf->pdev->dev, "HMC error interrupt\n");
+	}
+
 	/* If a critical error is pending we have no choice but to reset the
 	 * device.
 	 * Report and mask out any remaining unexpected interrupts.
@@ -2822,18 +2827,13 @@ static irqreturn_t i40e_intr(int irq, void *data)
 	if (icr0_remaining) {
 		dev_info(&pf->pdev->dev, "unhandled interrupt icr0=0x%08x\n",
 			 icr0_remaining);
-		if ((icr0_remaining & I40E_PFINT_ICR0_HMC_ERR_MASK) ||
-		    (icr0_remaining & I40E_PFINT_ICR0_PE_CRITERR_MASK) ||
+		if ((icr0_remaining & I40E_PFINT_ICR0_PE_CRITERR_MASK) ||
 		    (icr0_remaining & I40E_PFINT_ICR0_PCI_EXCEPTION_MASK) ||
 		    (icr0_remaining & I40E_PFINT_ICR0_ECC_ERR_MASK) ||
 		    (icr0_remaining & I40E_PFINT_ICR0_MAL_DETECT_MASK)) {
-			if (icr0 & I40E_PFINT_ICR0_HMC_ERR_MASK) {
-				dev_info(&pf->pdev->dev, "HMC error interrupt\n");
-			} else {
-				dev_info(&pf->pdev->dev, "device will be reset\n");
-				set_bit(__I40E_PF_RESET_REQUESTED, &pf->state);
-				i40e_service_event_schedule(pf);
-			}
+			dev_info(&pf->pdev->dev, "device will be reset\n");
+			set_bit(__I40E_PF_RESET_REQUESTED, &pf->state);
+			i40e_service_event_schedule(pf);
 		}
 		ena_mask &= ~icr0_remaining;
 	}
-- 
1.8.3.1

[net-next v3 03/16] i40e: Populate and check pci bus speed and width

[Jeff Kirsher]

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

Call i40e_set_pci_config_data from probe, then check that
we are in a 8GT/s x8 PCIe slot and send a warning if we are not.

Change-Id: I62815c574cee50d2787c50bbe956dde7a7a75a11
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_common.c    | 44 ++++++++++++++++++++++++
 drivers/net/ethernet/intel/i40e/i40e_main.c      | 23 +++++++++++++
 drivers/net/ethernet/intel/i40e/i40e_prototype.h |  1 +
 drivers/net/ethernet/intel/i40e/i40e_type.h      |  9 +++++
 4 files changed, 77 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 8b6d56a..cba1eb4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -2029,3 +2029,47 @@ i40e_status i40e_set_filter_control(struct i40e_hw *hw,
 
 	return 0;
 }
+/**
+ * i40e_set_pci_config_data - store PCI bus info
+ * @hw: pointer to hardware structure
+ * @link_status: the link status word from PCI config space
+ *
+ * Stores the PCI bus info (speed, width, type) within the i40e_hw structure
+ **/
+void i40e_set_pci_config_data(struct i40e_hw *hw, u16 link_status)
+{
+	hw->bus.type = i40e_bus_type_pci_express;
+
+	switch (link_status & PCI_EXP_LNKSTA_NLW) {
+	case I40E_PCI_LINK_WIDTH_1:
+		hw->bus.width = i40e_bus_width_pcie_x1;
+		break;
+	case I40E_PCI_LINK_WIDTH_2:
+		hw->bus.width = i40e_bus_width_pcie_x2;
+		break;
+	case I40E_PCI_LINK_WIDTH_4:
+		hw->bus.width = i40e_bus_width_pcie_x4;
+		break;
+	case I40E_PCI_LINK_WIDTH_8:
+		hw->bus.width = i40e_bus_width_pcie_x8;
+		break;
+	default:
+		hw->bus.width = i40e_bus_width_unknown;
+		break;
+	}
+
+	switch (link_status & PCI_EXP_LNKSTA_CLS) {
+	case I40E_PCI_LINK_SPEED_2500:
+		hw->bus.speed = i40e_bus_speed_2500;
+		break;
+	case I40E_PCI_LINK_SPEED_5000:
+		hw->bus.speed = i40e_bus_speed_5000;
+		break;
+	case I40E_PCI_LINK_SPEED_8000:
+		hw->bus.speed = i40e_bus_speed_8000;
+		break;
+	default:
+		hw->bus.speed = i40e_bus_speed_unknown;
+		break;
+	}
+}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index f14b31c..22a2c0e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -7369,6 +7369,7 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct i40e_pf *pf;
 	struct i40e_hw *hw;
 	static u16 pfs_found;
+	u16 link_status;
 	int err = 0;
 	u32 len;
 
@@ -7603,6 +7604,28 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	mod_timer(&pf->service_timer,
 		  round_jiffies(jiffies + pf->service_timer_period));
 
+	/* Get the negotiated link width and speed from PCI config space */
+	pcie_capability_read_word(pf->pdev, PCI_EXP_LNKSTA, &link_status);
+
+	i40e_set_pci_config_data(hw, link_status);
+
+	dev_info(&pdev->dev, "PCI Express: %s %s\n",
+		(hw->bus.speed == i40e_bus_speed_8000 ? "Speed 8.0GT/s" :
+		 hw->bus.speed == i40e_bus_speed_5000 ? "Speed 5.0GT/s" :
+		 hw->bus.speed == i40e_bus_speed_2500 ? "Speed 2.5GT/s" :
+		 "Unknown"),
+		(hw->bus.width == i40e_bus_width_pcie_x8 ? "Width x8" :
+		 hw->bus.width == i40e_bus_width_pcie_x4 ? "Width x4" :
+		 hw->bus.width == i40e_bus_width_pcie_x2 ? "Width x2" :
+		 hw->bus.width == i40e_bus_width_pcie_x1 ? "Width x1" :
+		 "Unknown"));
+
+	if (hw->bus.width < i40e_bus_width_pcie_x8 ||
+	    hw->bus.speed < i40e_bus_speed_8000) {
+		dev_warn(&pdev->dev, "PCI-Express bandwidth available for this device may be insufficient for optimal performance.\n");
+		dev_warn(&pdev->dev, "Please move the device to a different PCI-e link with more lanes and/or higher transfer rate.\n");
+	}
+
 	return 0;
 
 	/* Unwind what we've done if something failed in the setup */
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index 2fc9ce5..db7bf93 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -215,6 +215,7 @@ i40e_status i40e_read_nvm_buffer(struct i40e_hw *hw, u16 offset,
 					   u16 *words, u16 *data);
 i40e_status i40e_validate_nvm_checksum(struct i40e_hw *hw,
 						 u16 *checksum);
+void i40e_set_pci_config_data(struct i40e_hw *hw, u16 link_status);
 
 /* prototype for functions used for SW locks */
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index 72a6028..008a7a0 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -116,6 +116,15 @@ enum i40e_debug_mask {
 	I40E_DEBUG_ALL			= 0xFFFFFFFF
 };
 
+/* PCI Bus Info */
+#define I40E_PCI_LINK_WIDTH_1		0x10
+#define I40E_PCI_LINK_WIDTH_2		0x20
+#define I40E_PCI_LINK_WIDTH_4		0x40
+#define I40E_PCI_LINK_WIDTH_8		0x80
+#define I40E_PCI_LINK_SPEED_2500	0x1
+#define I40E_PCI_LINK_SPEED_5000	0x2
+#define I40E_PCI_LINK_SPEED_8000	0x3
+
 /* These are structs for managing the hardware information and the operations.
  * The structures of function pointers are filled out at init time when we
  * know for sure exactly which hardware we're working with.  This gives us the
-- 
1.8.3.1

[net-next v3 04/16] i40e: add wake-on-lan support

[Jeff Kirsher]

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

Wake on LAN is disabled by default and will remain that way for most
platforms, but there is an NVM setting that allows vendors to enable it
for a port if they think they've provided the right power environment
for the device.  This patch adds code to check the NVM setting and enable
Magic Packet use if WoL is enabled for the port.

Since only Magic Packet is supported, there's not a lot of HW configuration
needed.

Change-Id: I44e904a7b15695e34683009f487064cd86ea59b0
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@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_ethtool.c | 42 ++++++++++++++++++++++++--
 drivers/net/ethernet/intel/i40e/i40e_main.c    | 16 ++++++++--
 drivers/net/ethernet/intel/i40e/i40e_type.h    |  1 +
 4 files changed, 56 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 40706c1..31dd265 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -205,6 +205,7 @@ struct i40e_pf {
 	u16 rss_size_max;          /* HW defined max RSS queues */
 	u16 fdir_pf_filter_count;  /* num of guaranteed filters for this PF */
 	u8 atr_sample_rate;
+	bool wol_en;
 
 	enum i40e_interrupt_policy int_policy;
 	u16 rx_itr_default;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index d6681f6..255a138 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -844,8 +844,45 @@ static void i40e_diag_test(struct net_device *netdev,
 static void i40e_get_wol(struct net_device *netdev,
 			 struct ethtool_wolinfo *wol)
 {
-	wol->supported = 0;
-	wol->wolopts = 0;
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_pf *pf = np->vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	u16 wol_nvm_bits;
+
+	/* NVM bit on means WoL disabled for the port */
+	i40e_read_nvm_word(hw, I40E_SR_NVM_WAKE_ON_LAN, &wol_nvm_bits);
+	if ((1 << hw->port) & wol_nvm_bits) {
+		wol->supported = 0;
+		wol->wolopts = 0;
+	} else {
+		wol->supported = WAKE_MAGIC;
+		wol->wolopts = (pf->wol_en ? WAKE_MAGIC : 0);
+	}
+}
+
+static int i40e_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_pf *pf = np->vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	u16 wol_nvm_bits;
+
+	/* NVM bit on means WoL disabled for the port */
+	i40e_read_nvm_word(hw, I40E_SR_NVM_WAKE_ON_LAN, &wol_nvm_bits);
+	if (((1 << hw->port) & wol_nvm_bits))
+		return -EOPNOTSUPP;
+
+	/* only magic packet is supported */
+	if (wol->wolopts && (wol->wolopts != WAKE_MAGIC))
+		return -EOPNOTSUPP;
+
+	/* is this a new value? */
+	if (pf->wol_en != !!wol->wolopts) {
+		pf->wol_en = !!wol->wolopts;
+		device_set_wakeup_enable(&pf->pdev->dev, pf->wol_en);
+	}
+
+	return 0;
 }
 
 static int i40e_nway_reset(struct net_device *netdev)
@@ -1568,6 +1605,7 @@ static const struct ethtool_ops i40e_ethtool_ops = {
 	.nway_reset		= i40e_nway_reset,
 	.get_link		= ethtool_op_get_link,
 	.get_wol		= i40e_get_wol,
+	.set_wol		= i40e_set_wol,
 	.get_eeprom_len		= i40e_get_eeprom_len,
 	.get_eeprom		= i40e_get_eeprom,
 	.get_ringparam		= i40e_get_ringparam,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 22a2c0e..45b2198 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -7537,6 +7537,10 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	pf->flags |= I40E_FLAG_NEED_LINK_UPDATE;
 	pf->link_check_timeout = jiffies;
 
+	/* WoL defaults to disabled */
+	pf->wol_en = false;
+	device_set_wakeup_enable(&pf->pdev->dev, pf->wol_en);
+
 	/* set up the main switch operations */
 	i40e_determine_queue_usage(pf);
 	i40e_init_interrupt_scheme(pf);
@@ -7858,6 +7862,7 @@ static void i40e_pci_error_resume(struct pci_dev *pdev)
 static void i40e_shutdown(struct pci_dev *pdev)
 {
 	struct i40e_pf *pf = pci_get_drvdata(pdev);
+	struct i40e_hw *hw = &pf->hw;
 
 	set_bit(__I40E_SUSPENDED, &pf->state);
 	set_bit(__I40E_DOWN, &pf->state);
@@ -7865,8 +7870,11 @@ static void i40e_shutdown(struct pci_dev *pdev)
 	i40e_prep_for_reset(pf);
 	rtnl_unlock();
 
+	wr32(hw, I40E_PFPM_APM, (pf->wol_en ? I40E_PFPM_APM_APME_MASK : 0));
+	wr32(hw, I40E_PFPM_WUFC, (pf->wol_en ? I40E_PFPM_WUFC_MAG_MASK : 0));
+
 	if (system_state == SYSTEM_POWER_OFF) {
-		pci_wake_from_d3(pdev, false);    /* No WoL support yet */
+		pci_wake_from_d3(pdev, pf->wol_en);
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
 }
@@ -7879,6 +7887,7 @@ static void i40e_shutdown(struct pci_dev *pdev)
 static int i40e_suspend(struct pci_dev *pdev, pm_message_t state)
 {
 	struct i40e_pf *pf = pci_get_drvdata(pdev);
+	struct i40e_hw *hw = &pf->hw;
 
 	set_bit(__I40E_SUSPENDED, &pf->state);
 	set_bit(__I40E_DOWN, &pf->state);
@@ -7886,7 +7895,10 @@ static int i40e_suspend(struct pci_dev *pdev, pm_message_t state)
 	i40e_prep_for_reset(pf);
 	rtnl_unlock();
 
-	pci_wake_from_d3(pdev, false);    /* No WoL support yet */
+	wr32(hw, I40E_PFPM_APM, (pf->wol_en ? I40E_PFPM_APM_APME_MASK : 0));
+	wr32(hw, I40E_PFPM_WUFC, (pf->wol_en ? I40E_PFPM_WUFC_MAG_MASK : 0));
+
+	pci_wake_from_d3(pdev, pf->wol_en);
 	pci_set_power_state(pdev, PCI_D3hot);
 
 	return 0;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index 008a7a0..d683c30 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -1031,6 +1031,7 @@ struct i40e_hw_port_stats {
 #define I40E_SR_NVM_CONTROL_WORD		0x00
 #define I40E_SR_EMP_MODULE_PTR			0x0F
 #define I40E_SR_NVM_IMAGE_VERSION		0x18
+#define I40E_SR_NVM_WAKE_ON_LAN			0x19
 #define I40E_SR_ALTERNATE_SAN_MAC_ADDRESS_PTR	0x27
 #define I40E_SR_NVM_EETRACK_LO			0x2D
 #define I40E_SR_NVM_EETRACK_HI			0x2E
-- 
1.8.3.1

[net-next v3 05/16] i40e: fix curly brace use and return type

[Jeff Kirsher]

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

Add curly-braces on a multi-line function.  While we're here we
also change to return void in i40e_vsi_clear_rings() since no
caller cares.

Change-Id: I261fcef20e2a39e18d83ec08fdd14456131dee91
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 45b2198..b0cfb4c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5300,18 +5300,17 @@ free_vsi:
  * i40e_vsi_clear_rings - Deallocates the Rx and Tx rings for the provided VSI
  * @vsi: the VSI being cleaned
  **/
-static s32 i40e_vsi_clear_rings(struct i40e_vsi *vsi)
+static void i40e_vsi_clear_rings(struct i40e_vsi *vsi)
 {
 	int i;
 
-	if (vsi->tx_rings[0])
+	if (vsi->tx_rings[0]) {
 		for (i = 0; i < vsi->num_queue_pairs; i++) {
 			kfree_rcu(vsi->tx_rings[i], rcu);
 			vsi->tx_rings[i] = NULL;
 			vsi->rx_rings[i] = NULL;
 		}
-
-	return 0;
+	}
 }
 
 /**
-- 
1.8.3.1

[net-next v3 06/16] i40e: Implementation of vxlan ndo's

[Jeff Kirsher]

From: Joseph Gasparakis <joseph.gasparakis@intel.com>

This adds the implementation for the vxlan ndo's.  This allows
the hardware to do RX checksum offload for inner packets on the UDP
ports that vxlan notifies us about.

Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/Kconfig               |  1 +
 drivers/net/ethernet/intel/i40e/i40e.h           |  2 +
 drivers/net/ethernet/intel/i40e/i40e_common.c    | 57 +++++++++++++++
 drivers/net/ethernet/intel/i40e/i40e_main.c      | 92 ++++++++++++++++++++++++
 drivers/net/ethernet/intel/i40e/i40e_prototype.h |  6 ++
 drivers/net/ethernet/intel/i40e/i40e_type.h      |  1 +
 6 files changed, 159 insertions(+)

diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig
index 41a2fa2..38e47ed 100644
--- a/drivers/net/ethernet/intel/Kconfig
+++ b/drivers/net/ethernet/intel/Kconfig
@@ -244,6 +244,7 @@ config IXGBEVF
 config I40E
 	tristate "Intel(R) Ethernet Controller XL710 Family support"
 	depends on PCI
+	depends on VXLAN || !VXLAN
 	---help---
 	  This driver supports Intel(R) Ethernet Controller XL710 Family of
 	  devices.  For more information on how to identify your adapter, go
diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 31dd265..5d88cf4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -207,6 +207,8 @@ struct i40e_pf {
 	u8 atr_sample_rate;
 	bool wol_en;
 
+	__be16  vxlan_ports[I40E_MAX_PF_UDP_OFFLOAD_PORTS];
+
 	enum i40e_interrupt_policy int_policy;
 	u16 rx_itr_default;
 	u16 tx_itr_default;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index cba1eb4..ca58263 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -1671,6 +1671,63 @@ i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
 }
 
 /**
+ * i40e_aq_add_udp_tunnel
+ * @hw: pointer to the hw struct
+ * @udp_port: the UDP port to add
+ * @header_len: length of the tunneling header length in DWords
+ * @protocol_index: protocol index type
+ * @cmd_details: pointer to command details structure or NULL
+ **/
+i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
+				u16 udp_port, u8 header_len,
+				u8 protocol_index, u8 *filter_index,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_add_udp_tunnel *cmd =
+		(struct i40e_aqc_add_udp_tunnel *)&desc.params.raw;
+	struct i40e_aqc_del_udp_tunnel_completion *resp =
+		(struct i40e_aqc_del_udp_tunnel_completion *)&desc.params.raw;
+	i40e_status status;
+
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_udp_tunnel);
+
+	cmd->udp_port = cpu_to_le16(udp_port);
+	cmd->header_len = header_len;
+	cmd->protocol_index = protocol_index;
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	if (!status)
+		*filter_index = resp->index;
+
+	return status;
+}
+
+/**
+ * i40e_aq_del_udp_tunnel
+ * @hw: pointer to the hw struct
+ * @index: filter index
+ * @cmd_details: pointer to command details structure or NULL
+ **/
+i40e_status i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_remove_udp_tunnel *cmd =
+		(struct i40e_aqc_remove_udp_tunnel *)&desc.params.raw;
+	i40e_status status;
+
+	i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_del_udp_tunnel);
+
+	cmd->index = index;
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
  * i40e_aq_delete_element - Delete switch element
  * @hw: pointer to the hw struct
  * @seid: the SEID to delete from the switch
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index b0cfb4c..7b759a6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -27,6 +27,9 @@
 
 /* Local includes */
 #include "i40e.h"
+#if IS_ENABLED(CONFIG_VXLAN)
+#include "net/vxlan.h"
+#endif
 
 const char i40e_driver_name[] = "i40e";
 static const char i40e_driver_string[] =
@@ -3993,6 +3996,9 @@ static int i40e_open(struct net_device *netdev)
 				    "couldn't set broadcast err %d aq_err %d\n",
 				    err, pf->hw.aq.asq_last_status);
 	}
+#if IS_ENABLED(CONFIG_VXLAN)
+	vxlan_get_rx_port(netdev);
+#endif
 
 	return 0;
 
@@ -5900,6 +5906,90 @@ static int i40e_set_features(struct net_device *netdev,
 	return 0;
 }
 
+/**
+ * i40e_get_vxlan_port_idx - Lookup a possibly offloaded for Rx UDP port
+ * @pf: board private structure
+ * @port: The UDP port to look up
+ *
+ * Returns the index number or I40E_MAX_PF_UDP_OFFLOAD_PORTS if port not found
+ **/
+static u8 i40e_get_vxlan_port_idx(struct i40e_pf *pf, __be16 port)
+{
+	u8 i;
+
+	for (i = 0; i < I40E_MAX_PF_UDP_OFFLOAD_PORTS; i++) {
+		if (pf->vxlan_ports[i] == port)
+			return i;
+	}
+
+	return i;
+}
+
+/**
+ * i40e_add_vxlan_port - Get notifications about VXLAN ports that come up
+ * @netdev: This physical port's netdev
+ * @sa_family: Socket Family that VXLAN is notifying us about
+ * @port: New UDP port number that VXLAN started listening to
+ **/
+static void i40e_add_vxlan_port(struct net_device *netdev,
+				 sa_family_t sa_family, __be16 port)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	const int vxlan_hdr_qwords = 4;
+	struct i40e_hw *hw = &pf->hw;
+	i40e_status aq_ret;
+	u8 filter_index;
+
+	if (sa_family == AF_INET6)
+		return;
+
+	aq_ret = i40e_aq_add_udp_tunnel(hw, ntohs(port), vxlan_hdr_qwords,
+				     I40E_AQC_TUNNEL_TYPE_UDP, &filter_index,
+				     NULL);
+
+	if (!aq_ret && filter_index < I40E_MAX_PF_UDP_OFFLOAD_PORTS)
+		pf->vxlan_ports[filter_index] = port;
+	else
+		dev_warn(&pf->pdev->dev, "Could not use port %d for VXLAN offload\n",
+			 ntohs(port));
+}
+
+/**
+ * i40e_del_vxlan_port - Get notifications about VXLAN ports that go away
+ * @netdev: This physical port's netdev
+ * @sa_family: Socket Family that VXLAN is notifying us about
+ * @port: UDP port number that VXLAN stopped listening to
+ **/
+static void i40e_del_vxlan_port(struct net_device *netdev,
+				 sa_family_t sa_family, __be16 port)
+{
+	struct i40e_netdev_priv *np = netdev_priv(netdev);
+	struct i40e_vsi *vsi = np->vsi;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	i40e_status aq_ret;
+	u8 filter_index;
+
+	if (sa_family == AF_INET6)
+		return;
+
+	filter_index = i40e_get_vxlan_port_idx(pf, port);
+
+	if (filter_index < I40E_MAX_PF_UDP_OFFLOAD_PORTS) {
+		aq_ret = i40e_aq_del_udp_tunnel(hw, filter_index, NULL);
+		if (!aq_ret)
+			pf->vxlan_ports[filter_index] = 0;
+		else
+			dev_warn(&pf->pdev->dev, "Could not remove VXLAN port %d\n",
+				 ntohs(port));
+	} else {
+		dev_warn(&pf->pdev->dev, "Received notification to remove VXLAN port %d but port was not added before.\n",
+			 ntohs(port));
+	}
+}
+
 static const struct net_device_ops i40e_netdev_ops = {
 	.ndo_open		= i40e_open,
 	.ndo_stop		= i40e_close,
@@ -5921,6 +6011,8 @@ static const struct net_device_ops i40e_netdev_ops = {
 	.ndo_set_vf_vlan	= i40e_ndo_set_vf_port_vlan,
 	.ndo_set_vf_tx_rate	= i40e_ndo_set_vf_bw,
 	.ndo_get_vf_config	= i40e_ndo_get_vf_config,
+	.ndo_add_vxlan_port	= i40e_add_vxlan_port,
+	.ndo_del_vxlan_port	= i40e_del_vxlan_port,
 };
 
 /**
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index db7bf93..5c458bb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -157,6 +157,12 @@ i40e_status i40e_aq_stop_lldp(struct i40e_hw *hw, bool shutdown_agent,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
 				struct i40e_asq_cmd_details *cmd_details);
+i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
+				u16 udp_port, u8 header_len,
+				u8 protocol_index, u8 *filter_index,
+				struct i40e_asq_cmd_details *cmd_details);
+i40e_status i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
+				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_delete_element(struct i40e_hw *hw, u16 seid,
 				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_mac_address_write(struct i40e_hw *hw,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index d683c30..bb960cd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -59,6 +59,7 @@
 #define I40E_MAX_VSI_QP			16
 #define I40E_MAX_VF_VSI			3
 #define I40E_MAX_CHAINED_RX_BUFFERS	5
+#define I40E_MAX_PF_UDP_OFFLOAD_PORTS	16
 
 /* Max default timeout in ms, */
 #define I40E_MAX_NVM_TIMEOUT		18000
-- 
1.8.3.1

[net-next v3 07/16] i40e: Rx checksum offload for VXLAN

[Jeff Kirsher]

From: Joseph Gasparakis <joseph.gasparakis@intel.com>

This implements receive offload for VXLAN for i40e.  The hardware
supports checksum offload/verification of the inner/outer header.

Change-Id: I450db300af6713f2044fef1191a0d1d294c13369
Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@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 |  6 +--
 drivers/net/ethernet/intel/i40e/i40e_txrx.c | 57 ++++++++++++++++++++++++++---
 drivers/net/ethernet/intel/i40e/i40e_type.h | 51 ++++++++++++++------------
 4 files changed, 82 insertions(+), 33 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 5d88cf4..016b83d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -29,6 +29,7 @@
 #define _I40E_H_
 
 #include <net/tcp.h>
+#include <net/udp.h>
 #include <linux/init.h>
 #include <linux/types.h>
 #include <linux/errno.h>
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 7b759a6..a184074 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -27,9 +27,7 @@
 
 /* Local includes */
 #include "i40e.h"
-#if IS_ENABLED(CONFIG_VXLAN)
-#include "net/vxlan.h"
-#endif
+#include <net/vxlan.h>
 
 const char i40e_driver_name[] = "i40e";
 static const char i40e_driver_string[] =
@@ -3996,9 +3994,7 @@ static int i40e_open(struct net_device *netdev)
 				    "couldn't set broadcast err %d aq_err %d\n",
 				    err, pf->hw.aq.asq_last_status);
 	}
-#if IS_ENABLED(CONFIG_VXLAN)
 	vxlan_get_rx_port(netdev);
-#endif
 
 	return 0;
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 01d0334..a978451 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -860,12 +860,25 @@ static void i40e_receive_skb(struct i40e_ring *rx_ring,
  * @skb: skb currently being received and modified
  * @rx_status: status value of last descriptor in packet
  * @rx_error: error value of last descriptor in packet
+ * @rx_ptype: ptype value of last descriptor in packet
  **/
 static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 				    struct sk_buff *skb,
 				    u32 rx_status,
-				    u32 rx_error)
+				    u32 rx_error,
+				    u16 rx_ptype)
 {
+	bool ipv4_tunnel, ipv6_tunnel;
+	__wsum rx_udp_csum;
+	__sum16 csum;
+	struct iphdr *iph;
+
+	ipv4_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
+		      (rx_ptype < I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
+	ipv6_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT6_MAC_PAY3) &&
+		      (rx_ptype < I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4);
+
+	skb->encapsulation = ipv4_tunnel || ipv6_tunnel;
 	skb->ip_summed = CHECKSUM_NONE;
 
 	/* Rx csum enabled and ip headers found? */
@@ -873,13 +886,43 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	      rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
 		return;
 
-	/* IP or L4 checksum error */
+	/* 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_L4E_SHIFT) |
+			(1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))) {
 		vsi->back->hw_csum_rx_error++;
 		return;
 	}
 
+	if (ipv4_tunnel &&
+	    !(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);
+
+		/* Add 4 bytes for VLAN tagged packets */
+		skb->transport_header += (skb->protocol == htons(ETH_P_8021Q) ||
+					  skb->protocol == htons(ETH_P_8021AD))
+					  ? VLAN_HLEN : 0;
+
+		rx_udp_csum = udp_csum(skb);
+		iph = ip_hdr(skb);
+		csum = csum_tcpudp_magic(
+				iph->saddr, iph->daddr,
+				(skb->len - skb_transport_offset(skb)),
+				IPPROTO_UDP, rx_udp_csum);
+
+		if (udp_hdr(skb)->check != csum) {
+			vsi->back->hw_csum_rx_error++;
+			return;
+		}
+	}
+
 	skb->ip_summed = CHECKSUM_UNNECESSARY;
 }
 
@@ -920,6 +963,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
 	union i40e_rx_desc *rx_desc;
 	u32 rx_error, rx_status;
 	u64 qword;
+	u16 rx_ptype;
 
 	rx_desc = I40E_RX_DESC(rx_ring, i);
 	qword = le64_to_cpu(rx_desc->wb.qword1.status_error_len);
@@ -952,6 +996,8 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
 		rx_hbo = rx_error & (1 << I40E_RX_DESC_ERROR_HBO_SHIFT);
 		rx_error &= ~(1 << I40E_RX_DESC_ERROR_HBO_SHIFT);
 
+		rx_ptype = (qword & I40E_RXD_QW1_PTYPE_MASK) >>
+			   I40E_RXD_QW1_PTYPE_SHIFT;
 		rx_bi->skb = NULL;
 
 		/* This memory barrier is needed to keep us from reading
@@ -1032,13 +1078,14 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
 		}
 
 		skb->rxhash = i40e_rx_hash(rx_ring, rx_desc);
-		i40e_rx_checksum(vsi, skb, rx_status, rx_error);
-
 		/* probably a little skewed due to removing CRC */
 		total_rx_bytes += skb->len;
 		total_rx_packets++;
 
 		skb->protocol = eth_type_trans(skb, rx_ring->netdev);
+
+		i40e_rx_checksum(vsi, skb, rx_status, rx_error, rx_ptype);
+
 		vlan_tag = rx_status & (1 << I40E_RX_DESC_STATUS_L2TAG1P_SHIFT)
 			 ? le16_to_cpu(rx_desc->wb.qword0.lo_dword.l2tag1)
 			 : 0;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index bb960cd..6e5f35e7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -522,7 +522,8 @@ enum i40e_rx_desc_status_bits {
 	I40E_RX_DESC_STATUS_UMBCAST_SHIFT	= 9, /* 2 BITS */
 	I40E_RX_DESC_STATUS_FLM_SHIFT		= 11,
 	I40E_RX_DESC_STATUS_FLTSTAT_SHIFT	= 12, /* 2 BITS */
-	I40E_RX_DESC_STATUS_LPBK_SHIFT		= 14
+	I40E_RX_DESC_STATUS_LPBK_SHIFT		= 14,
+	I40E_RX_DESC_STATUS_UDP_0_SHIFT		= 16
 };
 
 #define I40E_RXD_QW1_STATUS_TSYNINDX_SHIFT   I40E_RX_DESC_STATUS_TSYNINDX_SHIFT
@@ -568,28 +569,32 @@ enum i40e_rx_desc_error_l3l4e_fcoe_masks {
 
 /* Packet type non-ip values */
 enum i40e_rx_l2_ptype {
-	I40E_RX_PTYPE_L2_RESERVED		= 0,
-	I40E_RX_PTYPE_L2_MAC_PAY2		= 1,
-	I40E_RX_PTYPE_L2_TIMESYNC_PAY2		= 2,
-	I40E_RX_PTYPE_L2_FIP_PAY2		= 3,
-	I40E_RX_PTYPE_L2_OUI_PAY2		= 4,
-	I40E_RX_PTYPE_L2_MACCNTRL_PAY2		= 5,
-	I40E_RX_PTYPE_L2_LLDP_PAY2		= 6,
-	I40E_RX_PTYPE_L2_ECP_PAY2		= 7,
-	I40E_RX_PTYPE_L2_EVB_PAY2		= 8,
-	I40E_RX_PTYPE_L2_QCN_PAY2		= 9,
-	I40E_RX_PTYPE_L2_EAPOL_PAY2		= 10,
-	I40E_RX_PTYPE_L2_ARP			= 11,
-	I40E_RX_PTYPE_L2_FCOE_PAY3		= 12,
-	I40E_RX_PTYPE_L2_FCOE_FCDATA_PAY3	= 13,
-	I40E_RX_PTYPE_L2_FCOE_FCRDY_PAY3	= 14,
-	I40E_RX_PTYPE_L2_FCOE_FCRSP_PAY3	= 15,
-	I40E_RX_PTYPE_L2_FCOE_FCOTHER_PA	= 16,
-	I40E_RX_PTYPE_L2_FCOE_VFT_PAY3		= 17,
-	I40E_RX_PTYPE_L2_FCOE_VFT_FCDATA	= 18,
-	I40E_RX_PTYPE_L2_FCOE_VFT_FCRDY		= 19,
-	I40E_RX_PTYPE_L2_FCOE_VFT_FCRSP		= 20,
-	I40E_RX_PTYPE_L2_FCOE_VFT_FCOTHER	= 21
+	I40E_RX_PTYPE_L2_RESERVED			= 0,
+	I40E_RX_PTYPE_L2_MAC_PAY2			= 1,
+	I40E_RX_PTYPE_L2_TIMESYNC_PAY2			= 2,
+	I40E_RX_PTYPE_L2_FIP_PAY2			= 3,
+	I40E_RX_PTYPE_L2_OUI_PAY2			= 4,
+	I40E_RX_PTYPE_L2_MACCNTRL_PAY2			= 5,
+	I40E_RX_PTYPE_L2_LLDP_PAY2			= 6,
+	I40E_RX_PTYPE_L2_ECP_PAY2			= 7,
+	I40E_RX_PTYPE_L2_EVB_PAY2			= 8,
+	I40E_RX_PTYPE_L2_QCN_PAY2			= 9,
+	I40E_RX_PTYPE_L2_EAPOL_PAY2			= 10,
+	I40E_RX_PTYPE_L2_ARP				= 11,
+	I40E_RX_PTYPE_L2_FCOE_PAY3			= 12,
+	I40E_RX_PTYPE_L2_FCOE_FCDATA_PAY3		= 13,
+	I40E_RX_PTYPE_L2_FCOE_FCRDY_PAY3		= 14,
+	I40E_RX_PTYPE_L2_FCOE_FCRSP_PAY3		= 15,
+	I40E_RX_PTYPE_L2_FCOE_FCOTHER_PA		= 16,
+	I40E_RX_PTYPE_L2_FCOE_VFT_PAY3			= 17,
+	I40E_RX_PTYPE_L2_FCOE_VFT_FCDATA		= 18,
+	I40E_RX_PTYPE_L2_FCOE_VFT_FCRDY			= 19,
+	I40E_RX_PTYPE_L2_FCOE_VFT_FCRSP			= 20,
+	I40E_RX_PTYPE_L2_FCOE_VFT_FCOTHER		= 21,
+	I40E_RX_PTYPE_GRENAT4_MAC_PAY3			= 58,
+	I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4	= 87,
+	I40E_RX_PTYPE_GRENAT6_MAC_PAY3			= 124,
+	I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4	= 153
 };
 
 struct i40e_rx_ptype_decoded {
-- 
1.8.3.1

[net-next v3 08/16] i40e: move i40e_reset_vf

[Jeff Kirsher]

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

The VF reset code will be refactored in future patches. Part of that
refactor required it to call i40e_alloc_vf_res and i40e_free_vf_res, so
the function must be moved. In order to make the future patches more
readable, we perform the function move here, with no other changes.

Change-Id: If6567c9c0bada6caafb2ee0227e0d9d50d05f27f
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 256 ++++++++++-----------
 1 file changed, 128 insertions(+), 128 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 1ea0886..c0e3aec 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -543,6 +543,134 @@ error_alloc_vsi_res:
 }
 
 /**
+ * i40e_enable_vf_mappings
+ * @vf: pointer to the vf info
+ *
+ * enable vf mappings
+ **/
+static void i40e_enable_vf_mappings(struct i40e_vf *vf)
+{
+	struct i40e_pf *pf = vf->pf;
+	struct i40e_hw *hw = &pf->hw;
+	u32 reg, total_queue_pairs = 0;
+	int j;
+
+	/* Tell the hardware we're using noncontiguous mapping. HW requires
+	 * that VF queues be mapped using this method, even when they are
+	 * contiguous in real life
+	 */
+	wr32(hw, I40E_VSILAN_QBASE(vf->lan_vsi_id),
+	     I40E_VSILAN_QBASE_VSIQTABLE_ENA_MASK);
+
+	/* enable VF vplan_qtable mappings */
+	reg = I40E_VPLAN_MAPENA_TXRX_ENA_MASK;
+	wr32(hw, I40E_VPLAN_MAPENA(vf->vf_id), reg);
+
+	/* map PF queues to VF queues */
+	for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
+		u16 qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index, j);
+		reg = (qid & I40E_VPLAN_QTABLE_QINDEX_MASK);
+		wr32(hw, I40E_VPLAN_QTABLE(total_queue_pairs, vf->vf_id), reg);
+		total_queue_pairs++;
+	}
+
+	/* map PF queues to VSI */
+	for (j = 0; j < 7; j++) {
+		if (j * 2 >= pf->vsi[vf->lan_vsi_index]->num_queue_pairs) {
+			reg = 0x07FF07FF;	/* unused */
+		} else {
+			u16 qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index,
+							  j * 2);
+			reg = qid;
+			qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index,
+						      (j * 2) + 1);
+			reg |= qid << 16;
+		}
+		wr32(hw, I40E_VSILAN_QTABLE(j, vf->lan_vsi_id), reg);
+	}
+
+	i40e_flush(hw);
+}
+
+/**
+ * i40e_disable_vf_mappings
+ * @vf: pointer to the vf info
+ *
+ * disable vf mappings
+ **/
+static void i40e_disable_vf_mappings(struct i40e_vf *vf)
+{
+	struct i40e_pf *pf = vf->pf;
+	struct i40e_hw *hw = &pf->hw;
+	int i;
+
+	/* disable qp mappings */
+	wr32(hw, I40E_VPLAN_MAPENA(vf->vf_id), 0);
+	for (i = 0; i < I40E_MAX_VSI_QP; i++)
+		wr32(hw, I40E_VPLAN_QTABLE(i, vf->vf_id),
+		     I40E_QUEUE_END_OF_LIST);
+	i40e_flush(hw);
+}
+
+/**
+ * i40e_free_vf_res
+ * @vf: pointer to the vf info
+ *
+ * free vf resources
+ **/
+static void i40e_free_vf_res(struct i40e_vf *vf)
+{
+	struct i40e_pf *pf = vf->pf;
+
+	/* free vsi & disconnect it from the parent uplink */
+	if (vf->lan_vsi_index) {
+		i40e_vsi_release(pf->vsi[vf->lan_vsi_index]);
+		vf->lan_vsi_index = 0;
+		vf->lan_vsi_id = 0;
+	}
+
+	/* reset some of the state varibles keeping
+	 * track of the resources
+	 */
+	vf->num_queue_pairs = 0;
+	vf->vf_states = 0;
+}
+
+/**
+ * i40e_alloc_vf_res
+ * @vf: pointer to the vf info
+ *
+ * allocate vf resources
+ **/
+static int i40e_alloc_vf_res(struct i40e_vf *vf)
+{
+	struct i40e_pf *pf = vf->pf;
+	int total_queue_pairs = 0;
+	int ret;
+
+	/* allocate hw vsi context & associated resources */
+	ret = i40e_alloc_vsi_res(vf, I40E_VSI_SRIOV);
+	if (ret)
+		goto error_alloc;
+	total_queue_pairs += pf->vsi[vf->lan_vsi_index]->num_queue_pairs;
+	set_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps);
+
+	/* store the total qps number for the runtime
+	 * vf req validation
+	 */
+	vf->num_queue_pairs = total_queue_pairs;
+
+	/* vf is now completely initialized */
+	set_bit(I40E_VF_STAT_INIT, &vf->vf_states);
+
+error_alloc:
+	if (ret)
+		i40e_free_vf_res(vf);
+
+	return ret;
+}
+
+/**
  * i40e_reset_vf
  * @vf: pointer to the vf structure
  * @flr: VFLR was issued or not
@@ -673,134 +801,6 @@ int i40e_reset_vf(struct i40e_vf *vf, bool flr)
 }
 
 /**
- * i40e_enable_vf_mappings
- * @vf: pointer to the vf info
- *
- * enable vf mappings
- **/
-static void i40e_enable_vf_mappings(struct i40e_vf *vf)
-{
-	struct i40e_pf *pf = vf->pf;
-	struct i40e_hw *hw = &pf->hw;
-	u32 reg, total_queue_pairs = 0;
-	int j;
-
-	/* Tell the hardware we're using noncontiguous mapping. HW requires
-	 * that VF queues be mapped using this method, even when they are
-	 * contiguous in real life
-	 */
-	wr32(hw, I40E_VSILAN_QBASE(vf->lan_vsi_id),
-	     I40E_VSILAN_QBASE_VSIQTABLE_ENA_MASK);
-
-	/* enable VF vplan_qtable mappings */
-	reg = I40E_VPLAN_MAPENA_TXRX_ENA_MASK;
-	wr32(hw, I40E_VPLAN_MAPENA(vf->vf_id), reg);
-
-	/* map PF queues to VF queues */
-	for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
-		u16 qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index, j);
-		reg = (qid & I40E_VPLAN_QTABLE_QINDEX_MASK);
-		wr32(hw, I40E_VPLAN_QTABLE(total_queue_pairs, vf->vf_id), reg);
-		total_queue_pairs++;
-	}
-
-	/* map PF queues to VSI */
-	for (j = 0; j < 7; j++) {
-		if (j * 2 >= pf->vsi[vf->lan_vsi_index]->num_queue_pairs) {
-			reg = 0x07FF07FF;	/* unused */
-		} else {
-			u16 qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index,
-							  j * 2);
-			reg = qid;
-			qid = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index,
-						      (j * 2) + 1);
-			reg |= qid << 16;
-		}
-		wr32(hw, I40E_VSILAN_QTABLE(j, vf->lan_vsi_id), reg);
-	}
-
-	i40e_flush(hw);
-}
-
-/**
- * i40e_disable_vf_mappings
- * @vf: pointer to the vf info
- *
- * disable vf mappings
- **/
-static void i40e_disable_vf_mappings(struct i40e_vf *vf)
-{
-	struct i40e_pf *pf = vf->pf;
-	struct i40e_hw *hw = &pf->hw;
-	int i;
-
-	/* disable qp mappings */
-	wr32(hw, I40E_VPLAN_MAPENA(vf->vf_id), 0);
-	for (i = 0; i < I40E_MAX_VSI_QP; i++)
-		wr32(hw, I40E_VPLAN_QTABLE(i, vf->vf_id),
-		     I40E_QUEUE_END_OF_LIST);
-	i40e_flush(hw);
-}
-
-/**
- * i40e_free_vf_res
- * @vf: pointer to the vf info
- *
- * free vf resources
- **/
-static void i40e_free_vf_res(struct i40e_vf *vf)
-{
-	struct i40e_pf *pf = vf->pf;
-
-	/* free vsi & disconnect it from the parent uplink */
-	if (vf->lan_vsi_index) {
-		i40e_vsi_release(pf->vsi[vf->lan_vsi_index]);
-		vf->lan_vsi_index = 0;
-		vf->lan_vsi_id = 0;
-	}
-
-	/* reset some of the state varibles keeping
-	 * track of the resources
-	 */
-	vf->num_queue_pairs = 0;
-	vf->vf_states = 0;
-}
-
-/**
- * i40e_alloc_vf_res
- * @vf: pointer to the vf info
- *
- * allocate vf resources
- **/
-static int i40e_alloc_vf_res(struct i40e_vf *vf)
-{
-	struct i40e_pf *pf = vf->pf;
-	int total_queue_pairs = 0;
-	int ret;
-
-	/* allocate hw vsi context & associated resources */
-	ret = i40e_alloc_vsi_res(vf, I40E_VSI_SRIOV);
-	if (ret)
-		goto error_alloc;
-	total_queue_pairs += pf->vsi[vf->lan_vsi_index]->num_queue_pairs;
-	set_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps);
-
-	/* store the total qps number for the runtime
-	 * vf req validation
-	 */
-	vf->num_queue_pairs = total_queue_pairs;
-
-	/* vf is now completely initialized */
-	set_bit(I40E_VF_STAT_INIT, &vf->vf_states);
-
-error_alloc:
-	if (ret)
-		i40e_free_vf_res(vf);
-
-	return ret;
-}
-
-/**
  * i40e_vfs_are_assigned
  * @pf: pointer to the pf structure
  *
-- 
1.8.3.1

[net-next v3 09/16] i40e: refactor VF reset flow

[Jeff Kirsher]

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

Fix the VF reset flow so that it works on real hardware. After
discussions with the HW team, the reset flow has been changed
somewhat.

- Change the i40e_reset_vf function to a void type, and fix
  up the callers to reflect this.
- Move the MSI-X disable code to i40e_free_vf_res since it must
  be done every time the VF is freed, regardless of whether or
  not it is reset.
- Ensure that the PCIe bus is quiet before polling the reset bit.
- Don't clear the VFGEN_RSTAT1 register at the beginning as it is
  cleared by the reset.
- Poll longer for the reset to be done.
- Disable the queues using an existing function rather than
  rolling our own.
- Free and reallocate the VSI after reset to avoid rx hang.

Change-Id: I11e2590431cb73e8663714d1cc5b23d59b809033
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@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        |   2 +-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 202 ++++++++++-----------
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h |   2 +-
 4 files changed, 99 insertions(+), 108 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 016b83d..276eb3e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -537,6 +537,7 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type,
 int i40e_vsi_release(struct i40e_vsi *vsi);
 struct i40e_vsi *i40e_vsi_lookup(struct i40e_pf *pf, enum i40e_vsi_type type,
 				 struct i40e_vsi *start_vsi);
+int i40e_vsi_control_rings(struct i40e_vsi *vsi, bool enable);
 int i40e_reconfig_rss_queues(struct i40e_pf *pf, int queue_count);
 struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags, u16 uplink_seid,
 				u16 downlink_seid, u8 enabled_tc);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index a184074..c0f78bc 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3108,7 +3108,7 @@ static int i40e_vsi_control_rx(struct i40e_vsi *vsi, bool enable)
  * @vsi: the VSI being configured
  * @enable: start or stop the rings
  **/
-static int i40e_vsi_control_rings(struct i40e_vsi *vsi, bool request)
+int i40e_vsi_control_rings(struct i40e_vsi *vsi, bool request)
 {
 	int ret;
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index c0e3aec..8fdc842 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -621,6 +621,9 @@ static void i40e_disable_vf_mappings(struct i40e_vf *vf)
 static void i40e_free_vf_res(struct i40e_vf *vf)
 {
 	struct i40e_pf *pf = vf->pf;
+	struct i40e_hw *hw = &pf->hw;
+	u32 reg_idx, reg;
+	int i, msix_vf;
 
 	/* free vsi & disconnect it from the parent uplink */
 	if (vf->lan_vsi_index) {
@@ -628,7 +631,34 @@ static void i40e_free_vf_res(struct i40e_vf *vf)
 		vf->lan_vsi_index = 0;
 		vf->lan_vsi_id = 0;
 	}
+	msix_vf = pf->hw.func_caps.num_msix_vectors_vf + 1;
+	/* disable interrupts so the VF starts in a known state */
+	for (i = 0; i < msix_vf; i++) {
+		/* format is same for both registers */
+		if (0 == i)
+			reg_idx = I40E_VFINT_DYN_CTL0(vf->vf_id);
+		else
+			reg_idx = I40E_VFINT_DYN_CTLN(((msix_vf - 1) *
+						      (vf->vf_id))
+						     + (i - 1));
+		wr32(hw, reg_idx, I40E_VFINT_DYN_CTLN_CLEARPBA_MASK);
+		i40e_flush(hw);
+	}
 
+	/* clear the irq settings */
+	for (i = 0; i < msix_vf; i++) {
+		/* format is same for both registers */
+		if (0 == i)
+			reg_idx = I40E_VPINT_LNKLST0(vf->vf_id);
+		else
+			reg_idx = I40E_VPINT_LNKLSTN(((msix_vf - 1) *
+						      (vf->vf_id))
+						     + (i - 1));
+		reg = (I40E_VPINT_LNKLSTN_FIRSTQ_TYPE_MASK |
+		       I40E_VPINT_LNKLSTN_FIRSTQ_INDX_MASK);
+		wr32(hw, reg_idx, reg);
+		i40e_flush(hw);
+	}
 	/* reset some of the state varibles keeping
 	 * track of the resources
 	 */
@@ -670,6 +700,36 @@ error_alloc:
 	return ret;
 }
 
+#define VF_DEVICE_STATUS 0xAA
+#define VF_TRANS_PENDING_MASK 0x20
+/**
+ * i40e_quiesce_vf_pci
+ * @vf: pointer to the vf structure
+ *
+ * Wait for VF PCI transactions to be cleared after reset. Returns -EIO
+ * if the transactions never clear.
+ **/
+static int i40e_quiesce_vf_pci(struct i40e_vf *vf)
+{
+	struct i40e_pf *pf = vf->pf;
+	struct i40e_hw *hw = &pf->hw;
+	int vf_abs_id, i;
+	u32 reg;
+
+	reg = rd32(hw, I40E_PF_VT_PFALLOC);
+	vf_abs_id = vf->vf_id + (reg & I40E_PF_VT_PFALLOC_FIRSTVF_MASK);
+
+	wr32(hw, I40E_PF_PCI_CIAA,
+	     VF_DEVICE_STATUS | (vf_abs_id << I40E_PF_PCI_CIAA_VF_NUM_SHIFT));
+	for (i = 0; i < 100; i++) {
+		reg = rd32(hw, I40E_PF_PCI_CIAD);
+		if ((reg & VF_TRANS_PENDING_MASK) == 0)
+			return 0;
+		udelay(1);
+	}
+	return -EIO;
+}
+
 /**
  * i40e_reset_vf
  * @vf: pointer to the vf structure
@@ -677,35 +737,36 @@ error_alloc:
  *
  * reset the vf
  **/
-int i40e_reset_vf(struct i40e_vf *vf, bool flr)
+void i40e_reset_vf(struct i40e_vf *vf, bool flr)
 {
-	int ret = -ENOENT;
 	struct i40e_pf *pf = vf->pf;
 	struct i40e_hw *hw = &pf->hw;
-	u32 reg, reg_idx, msix_vf;
 	bool rsd = false;
-	u16 pf_queue_id;
-	int i, j;
+	int i;
+	u32 reg;
 
 	/* warn the VF */
-	wr32(hw, I40E_VFGEN_RSTAT1(vf->vf_id), I40E_VFR_INPROGRESS);
-
 	clear_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states);
 
-	/* PF triggers VFR only when VF requests, in case of
-	 * VFLR, HW triggers VFR
+	/* In the case of a VFLR, the HW has already reset the VF and we
+	 * just need to clean up, so don't hit the VFRTRIG register.
 	 */
 	if (!flr) {
 		/* reset vf using VPGEN_VFRTRIG reg */
-		reg = I40E_VPGEN_VFRTRIG_VFSWR_MASK;
+		reg = rd32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id));
+		reg |= I40E_VPGEN_VFRTRIG_VFSWR_MASK;
 		wr32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id), reg);
 		i40e_flush(hw);
 	}
 
+	if (i40e_quiesce_vf_pci(vf))
+		dev_err(&pf->pdev->dev, "VF %d PCI transactions stuck\n",
+			vf->vf_id);
+
 	/* poll VPGEN_VFRSTAT reg to make sure
 	 * that reset is complete
 	 */
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < 100; i++) {
 		/* vf reset requires driver to first reset the
 		 * vf & than poll the status register to make sure
 		 * that the requested op was completed
@@ -720,84 +781,29 @@ int i40e_reset_vf(struct i40e_vf *vf, bool flr)
 	}
 
 	if (!rsd)
-		dev_err(&pf->pdev->dev, "VF reset check timeout %d\n",
+		dev_err(&pf->pdev->dev, "VF reset check timeout on VF %d\n",
 			vf->vf_id);
-
-	/* fast disable qps */
-	for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
-		ret = i40e_ctrl_vsi_tx_queue(vf, vf->lan_vsi_index, j,
-					     I40E_QUEUE_CTRL_FASTDISABLE);
-		ret = i40e_ctrl_vsi_rx_queue(vf, vf->lan_vsi_index, j,
-					     I40E_QUEUE_CTRL_FASTDISABLE);
-	}
-
-	/* Queue enable/disable requires driver to
-	 * first reset the vf & than poll the status register
-	 * to make sure that the requested op was completed
-	 * successfully
-	 */
-	udelay(10);
-	for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
-		ret = i40e_ctrl_vsi_tx_queue(vf, vf->lan_vsi_index, j,
-					     I40E_QUEUE_CTRL_FASTDISABLECHECK);
-		if (ret)
-			dev_info(&pf->pdev->dev,
-				 "Queue control check failed on Tx queue %d of VSI %d VF %d\n",
-				 j, vf->lan_vsi_index, vf->vf_id);
-		ret = i40e_ctrl_vsi_rx_queue(vf, vf->lan_vsi_index, j,
-					     I40E_QUEUE_CTRL_FASTDISABLECHECK);
-		if (ret)
-			dev_info(&pf->pdev->dev,
-				 "Queue control check failed on Rx queue %d of VSI %d VF %d\n",
-				 j, vf->lan_vsi_index, vf->vf_id);
-	}
-
-	/* clear the irq settings */
-	msix_vf = pf->hw.func_caps.num_msix_vectors_vf;
-	for (i = 0; i < msix_vf; i++) {
-		/* format is same for both registers */
-		if (0 == i)
-			reg_idx = I40E_VPINT_LNKLST0(vf->vf_id);
-		else
-			reg_idx = I40E_VPINT_LNKLSTN(((msix_vf - 1) *
-						      (vf->vf_id))
-						     + (i - 1));
-		reg = (I40E_VPINT_LNKLSTN_FIRSTQ_TYPE_MASK |
-		       I40E_VPINT_LNKLSTN_FIRSTQ_INDX_MASK);
-		wr32(hw, reg_idx, reg);
-		i40e_flush(hw);
-	}
-	/* disable interrupts so the VF starts in a known state */
-	for (i = 0; i < msix_vf; i++) {
-		/* format is same for both registers */
-		if (0 == i)
-			reg_idx = I40E_VFINT_DYN_CTL0(vf->vf_id);
-		else
-			reg_idx = I40E_VFINT_DYN_CTLN(((msix_vf - 1) *
-						      (vf->vf_id))
-						     + (i - 1));
-		wr32(hw, reg_idx, I40E_VFINT_DYN_CTLN_CLEARPBA_MASK);
-		i40e_flush(hw);
-	}
-
-	/* set the defaults for the rqctl & tqctl registers */
-	reg = (I40E_QINT_RQCTL_NEXTQ_INDX_MASK | I40E_QINT_RQCTL_ITR_INDX_MASK |
-	       I40E_QINT_RQCTL_NEXTQ_TYPE_MASK);
-	for (j = 0; j < pf->vsi[vf->lan_vsi_index]->num_queue_pairs; j++) {
-		pf_queue_id = i40e_vc_get_pf_queue_id(vf, vf->lan_vsi_index, j);
-		wr32(hw, I40E_QINT_RQCTL(pf_queue_id), reg);
-		wr32(hw, I40E_QINT_TQCTL(pf_queue_id), reg);
-	}
-
+	wr32(hw, I40E_VFGEN_RSTAT1(vf->vf_id), I40E_VFR_COMPLETED);
 	/* clear the reset bit in the VPGEN_VFRTRIG reg */
 	reg = rd32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id));
 	reg &= ~I40E_VPGEN_VFRTRIG_VFSWR_MASK;
 	wr32(hw, I40E_VPGEN_VFRTRIG(vf->vf_id), reg);
+
+	/* On initial reset, we won't have any queues */
+	if (vf->lan_vsi_index == 0)
+		goto complete_reset;
+
+	i40e_vsi_control_rings(pf->vsi[vf->lan_vsi_index], false);
+complete_reset:
+	/* reallocate vf resources to reset the VSI state */
+	i40e_free_vf_res(vf);
+	mdelay(10);
+	i40e_alloc_vf_res(vf);
+	i40e_enable_vf_mappings(vf);
+
 	/* tell the VF the reset is done */
-	wr32(hw, I40E_VFGEN_RSTAT1(vf->vf_id), I40E_VFR_COMPLETED);
+	wr32(hw, I40E_VFGEN_RSTAT1(vf->vf_id), I40E_VFR_VFACTIVE);
 	i40e_flush(hw);
-
-	return ret;
 }
 
 /**
@@ -909,11 +915,8 @@ static 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);
-
-		ret = i40e_alloc_vf_res(&vfs[i]);
-		i40e_reset_vf(&vfs[i], true);
-		if (ret)
-			break;
+		/* vf resources get allocated during reset */
+		i40e_reset_vf(&vfs[i], false);
 
 		/* enable vf vplan_qtable mappings */
 		i40e_enable_vf_mappings(&vfs[i]);
@@ -1140,12 +1143,10 @@ err:
  * unlike other virtchnl messages, pf driver
  * doesn't send the response back to the vf
  **/
-static int i40e_vc_reset_vf_msg(struct i40e_vf *vf)
+static void i40e_vc_reset_vf_msg(struct i40e_vf *vf)
 {
-	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states))
-		return -ENOENT;
-
-	return i40e_reset_vf(vf, false);
+	if (test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states))
+		i40e_reset_vf(vf, false);
 }
 
 /**
@@ -1879,7 +1880,8 @@ int i40e_vc_process_vf_msg(struct i40e_pf *pf, u16 vf_id, u32 v_opcode,
 		ret = i40e_vc_get_vf_resources_msg(vf);
 		break;
 	case I40E_VIRTCHNL_OP_RESET_VF:
-		ret = i40e_vc_reset_vf_msg(vf);
+		i40e_vc_reset_vf_msg(vf);
+		ret = 0;
 		break;
 	case I40E_VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE:
 		ret = i40e_vc_config_promiscuous_mode_msg(vf, msg, msglen);
@@ -1950,19 +1952,7 @@ int i40e_vc_process_vflr_event(struct i40e_pf *pf)
 			/* clear the bit in GLGEN_VFLRSTAT */
 			wr32(hw, I40E_GLGEN_VFLRSTAT(reg_idx), (1 << bit_idx));
 
-			if (i40e_reset_vf(vf, true))
-				dev_err(&pf->pdev->dev,
-					"Unable to reset the VF %d\n", vf_id);
-			/* free up vf resources to destroy vsi state */
-			i40e_free_vf_res(vf);
-
-			/* allocate new vf resources with the default state */
-			if (i40e_alloc_vf_res(vf))
-				dev_err(&pf->pdev->dev,
-					"Unable to allocate VF resources %d\n",
-					vf_id);
-
-			i40e_enable_vf_mappings(vf);
+			i40e_reset_vf(vf, true);
 		}
 	}
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
index 360382c..d0b712c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
@@ -104,7 +104,7 @@ int i40e_pci_sriov_configure(struct pci_dev *dev, int num_vfs);
 int i40e_vc_process_vf_msg(struct i40e_pf *pf, u16 vf_id, u32 v_opcode,
 			   u32 v_retval, u8 *msg, u16 msglen);
 int i40e_vc_process_vflr_event(struct i40e_pf *pf);
-int i40e_reset_vf(struct i40e_vf *vf, bool flr);
+void i40e_reset_vf(struct i40e_vf *vf, bool flr);
 void i40e_vc_notify_vf_reset(struct i40e_vf *vf);
 
 /* vf configuration related iplink handlers */
-- 
1.8.3.1

[net-next v3 10/16] i40e: remove redundant code

[Jeff Kirsher]

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

Don't keep separate functions to enable and disable queues for the VFs.
Just call the existing function that everybody else uses. Remove the
unused functions.

Change-Id: I15db9aad64a59e502bfe1e0fdab9b347ab85c12c
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 223 +--------------------
 1 file changed, 4 insertions(+), 219 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 8fdc842..1a7052a 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -102,130 +102,6 @@ static u16 i40e_vc_get_pf_queue_id(struct i40e_vf *vf, u8 vsi_idx,
 }
 
 /**
- * i40e_ctrl_vsi_tx_queue
- * @vf: pointer to the vf info
- * @vsi_idx: index of VSI in PF struct
- * @vsi_queue_id: vsi relative queue index
- * @ctrl: control flags
- *
- * enable/disable/enable check/disable check
- **/
-static int i40e_ctrl_vsi_tx_queue(struct i40e_vf *vf, u16 vsi_idx,
-				  u16 vsi_queue_id,
-				  enum i40e_queue_ctrl ctrl)
-{
-	struct i40e_pf *pf = vf->pf;
-	struct i40e_hw *hw = &pf->hw;
-	bool writeback = false;
-	u16 pf_queue_id;
-	int ret = 0;
-	u32 reg;
-
-	pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx, vsi_queue_id);
-	reg = rd32(hw, I40E_QTX_ENA(pf_queue_id));
-
-	switch (ctrl) {
-	case I40E_QUEUE_CTRL_ENABLE:
-		reg |= I40E_QTX_ENA_QENA_REQ_MASK;
-		writeback = true;
-		break;
-	case I40E_QUEUE_CTRL_ENABLECHECK:
-		ret = (reg & I40E_QTX_ENA_QENA_STAT_MASK) ? 0 : -EPERM;
-		break;
-	case I40E_QUEUE_CTRL_DISABLE:
-		reg &= ~I40E_QTX_ENA_QENA_REQ_MASK;
-		writeback = true;
-		break;
-	case I40E_QUEUE_CTRL_DISABLECHECK:
-		ret = (reg & I40E_QTX_ENA_QENA_STAT_MASK) ? -EPERM : 0;
-		break;
-	case I40E_QUEUE_CTRL_FASTDISABLE:
-		reg |= I40E_QTX_ENA_FAST_QDIS_MASK;
-		writeback = true;
-		break;
-	case I40E_QUEUE_CTRL_FASTDISABLECHECK:
-		ret = (reg & I40E_QTX_ENA_QENA_STAT_MASK) ? -EPERM : 0;
-		if (!ret) {
-			reg &= ~I40E_QTX_ENA_FAST_QDIS_MASK;
-			writeback = true;
-		}
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	}
-
-	if (writeback) {
-		wr32(hw, I40E_QTX_ENA(pf_queue_id), reg);
-		i40e_flush(hw);
-	}
-
-	return ret;
-}
-
-/**
- * i40e_ctrl_vsi_rx_queue
- * @vf: pointer to the vf info
- * @vsi_idx: index of VSI in PF struct
- * @vsi_queue_id: vsi relative queue index
- * @ctrl: control flags
- *
- * enable/disable/enable check/disable check
- **/
-static int i40e_ctrl_vsi_rx_queue(struct i40e_vf *vf, u16 vsi_idx,
-				  u16 vsi_queue_id,
-				  enum i40e_queue_ctrl ctrl)
-{
-	struct i40e_pf *pf = vf->pf;
-	struct i40e_hw *hw = &pf->hw;
-	bool writeback = false;
-	u16 pf_queue_id;
-	int ret = 0;
-	u32 reg;
-
-	pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_idx, vsi_queue_id);
-	reg = rd32(hw, I40E_QRX_ENA(pf_queue_id));
-
-	switch (ctrl) {
-	case I40E_QUEUE_CTRL_ENABLE:
-		reg |= I40E_QRX_ENA_QENA_REQ_MASK;
-		writeback = true;
-		break;
-	case I40E_QUEUE_CTRL_ENABLECHECK:
-		ret = (reg & I40E_QRX_ENA_QENA_STAT_MASK) ? 0 : -EPERM;
-		break;
-	case I40E_QUEUE_CTRL_DISABLE:
-		reg &= ~I40E_QRX_ENA_QENA_REQ_MASK;
-		writeback = true;
-		break;
-	case I40E_QUEUE_CTRL_DISABLECHECK:
-		ret = (reg & I40E_QRX_ENA_QENA_STAT_MASK) ? -EPERM : 0;
-		break;
-	case I40E_QUEUE_CTRL_FASTDISABLE:
-		reg |= I40E_QRX_ENA_FAST_QDIS_MASK;
-		writeback = true;
-		break;
-	case I40E_QUEUE_CTRL_FASTDISABLECHECK:
-		ret = (reg & I40E_QRX_ENA_QENA_STAT_MASK) ? -EPERM : 0;
-		if (!ret) {
-			reg &= ~I40E_QRX_ENA_FAST_QDIS_MASK;
-			writeback = true;
-		}
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	}
-
-	if (writeback) {
-		wr32(hw, I40E_QRX_ENA(pf_queue_id), reg);
-		i40e_flush(hw);
-	}
-
-	return ret;
-}
-
-/**
  * i40e_config_irq_link_list
  * @vf: pointer to the vf info
  * @vsi_idx: index of VSI in PF struct
@@ -1328,8 +1204,6 @@ static int i40e_vc_enable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	struct i40e_pf *pf = vf->pf;
 	u16 vsi_id = vqs->vsi_id;
 	i40e_status aq_ret = 0;
-	unsigned long tempmap;
-	u16 queue_id;
 
 	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
 		aq_ret = I40E_ERR_PARAM;
@@ -1345,52 +1219,8 @@ static int i40e_vc_enable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 		aq_ret = I40E_ERR_PARAM;
 		goto error_param;
 	}
-
-	tempmap = vqs->rx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
-			aq_ret = I40E_ERR_PARAM;
-			goto error_param;
-		}
-		i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
-				       I40E_QUEUE_CTRL_ENABLE);
-	}
-
-	tempmap = vqs->tx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
-			aq_ret = I40E_ERR_PARAM;
-			goto error_param;
-		}
-		i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
-				       I40E_QUEUE_CTRL_ENABLE);
-	}
-
-	/* Poll the status register to make sure that the
-	 * requested op was completed successfully
-	 */
-	udelay(10);
-
-	tempmap = vqs->rx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
-					   I40E_QUEUE_CTRL_ENABLECHECK)) {
-			dev_err(&pf->pdev->dev,
-				"Queue control check failed on RX queue %d of VSI %d VF %d\n",
-				queue_id, vsi_id, vf->vf_id);
-		}
-	}
-
-	tempmap = vqs->tx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
-					   I40E_QUEUE_CTRL_ENABLECHECK)) {
-			dev_err(&pf->pdev->dev,
-				"Queue control check failed on TX queue %d of VSI %d VF %d\n",
-				queue_id, vsi_id, vf->vf_id);
-		}
-	}
-
+	if (i40e_vsi_control_rings(pf->vsi[vsi_id], true))
+		aq_ret = I40E_ERR_TIMEOUT;
 error_param:
 	/* send the response to the vf */
 	return i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_ENABLE_QUEUES,
@@ -1413,8 +1243,6 @@ static int i40e_vc_disable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	struct i40e_pf *pf = vf->pf;
 	u16 vsi_id = vqs->vsi_id;
 	i40e_status aq_ret = 0;
-	unsigned long tempmap;
-	u16 queue_id;
 
 	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states)) {
 		aq_ret = I40E_ERR_PARAM;
@@ -1430,51 +1258,8 @@ static int i40e_vc_disable_queues_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 		aq_ret = I40E_ERR_PARAM;
 		goto error_param;
 	}
-
-	tempmap = vqs->rx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
-			aq_ret = I40E_ERR_PARAM;
-			goto error_param;
-		}
-		i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
-				       I40E_QUEUE_CTRL_DISABLE);
-	}
-
-	tempmap = vqs->tx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (!i40e_vc_isvalid_queue_id(vf, vsi_id, queue_id)) {
-			aq_ret = I40E_ERR_PARAM;
-			goto error_param;
-		}
-		i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
-				       I40E_QUEUE_CTRL_DISABLE);
-	}
-
-	/* Poll the status register to make sure that the
-	 * requested op was completed successfully
-	 */
-	udelay(10);
-
-	tempmap = vqs->rx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (i40e_ctrl_vsi_rx_queue(vf, vsi_id, queue_id,
-					   I40E_QUEUE_CTRL_DISABLECHECK)) {
-			dev_err(&pf->pdev->dev,
-				"Queue control check failed on RX queue %d of VSI %d VF %d\n",
-				queue_id, vsi_id, vf->vf_id);
-		}
-	}
-
-	tempmap = vqs->tx_queues;
-	for_each_set_bit(queue_id, &tempmap, I40E_MAX_VSI_QP) {
-		if (i40e_ctrl_vsi_tx_queue(vf, vsi_id, queue_id,
-					   I40E_QUEUE_CTRL_DISABLECHECK)) {
-			dev_err(&pf->pdev->dev,
-				"Queue control check failed on TX queue %d of VSI %d VF %d\n",
-				queue_id, vsi_id, vf->vf_id);
-		}
-	}
+	if (i40e_vsi_control_rings(pf->vsi[vsi_id], false))
+		aq_ret = I40E_ERR_TIMEOUT;
 
 error_param:
 	/* send the response to the vf */
-- 
1.8.3.1

[net-next v3 11/16] i40e: remove chatty log messages

[Jeff Kirsher]

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

Don't complain when we disable queues that are already disable, or
enable them when they're already enabled. This removes a bunch of bogus
log messages that we see at every VF reset.

Change-Id: Ia127be572abdccc48a53d8c43f8a07b8bb920de1
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 20 +++++---------------
 1 file changed, 5 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index c0f78bc..578f218 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -2988,21 +2988,11 @@ static int i40e_vsi_control_tx(struct i40e_vsi *vsi, bool enable)
 		} while (j-- && ((tx_reg >> I40E_QTX_ENA_QENA_REQ_SHIFT)
 			       ^ (tx_reg >> I40E_QTX_ENA_QENA_STAT_SHIFT)) & 1);
 
-		if (enable) {
-			/* is STAT set ? */
-			if ((tx_reg & I40E_QTX_ENA_QENA_STAT_MASK)) {
-				dev_info(&pf->pdev->dev,
-					 "Tx %d already enabled\n", i);
-				continue;
-			}
-		} else {
-			/* is !STAT set ? */
-			if (!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK)) {
-				dev_info(&pf->pdev->dev,
-					 "Tx %d already disabled\n", i);
-				continue;
-			}
-		}
+		/* Skip if the queue is already in the requested state */
+		if (enable && (tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
+			continue;
+		if (!enable && !(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
+			continue;
 
 		/* turn on/off the queue */
 		if (enable)
-- 
1.8.3.1

[net-next v3 12/16] i40e: fix error return

[Jeff Kirsher]

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

If the VF asks to add an invalid MAC address, tell it that instead of
just using a generic return code.

Change-Id: I366aff5449fa5874ad51e2734cac2a71783ab14b
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 1a7052a..4386a2f 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1340,7 +1340,7 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 		    is_zero_ether_addr(al->list[i].addr)) {
 			dev_err(&pf->pdev->dev, "invalid VF MAC addr %pMAC\n",
 				al->list[i].addr);
-			aq_ret = I40E_ERR_PARAM;
+			aq_ret = I40E_ERR_INVALID_MAC_ADDR;
 			goto error_param;
 		}
 	}
-- 
1.8.3.1

[net-next v3 13/16] i40e: be more informative

[Jeff Kirsher]

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

If the VF sends a bad message, be more informative about what it
actually is.

Change-Id: I89e06d2db416a1d05aeea016dd6e8b7870cae99a
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 4386a2f..1dd13f4 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1653,7 +1653,8 @@ int i40e_vc_process_vf_msg(struct i40e_pf *pf, u16 vf_id, u32 v_opcode,
 	ret = i40e_vc_validate_vf_msg(vf, v_opcode, v_retval, msg, msglen);
 
 	if (ret) {
-		dev_err(&pf->pdev->dev, "invalid message from vf %d\n", vf_id);
+		dev_err(&pf->pdev->dev, "Invalid message from vf %d, opcode %d, len %d\n",
+			vf_id, v_opcode, msglen);
 		return ret;
 	}
 	wr32(hw, I40E_VFGEN_RSTAT1(vf_id), I40E_VFR_VFACTIVE);
-- 
1.8.3.1

[net-next v3 14/16] i40e: make a define from a large constant

[Jeff Kirsher]

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

Make a define used in the header file by both VF and PF drivers.

Change-Id: Ie9e35adcc021cd6a8f7513934984eb4ed55774f5
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 21 +--------------------
 drivers/net/ethernet/intel/i40e/i40e_txrx.h | 20 ++++++++++++++++++++
 2 files changed, 21 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 578f218..aec2483 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5650,25 +5650,6 @@ static int i40e_setup_misc_vector(struct i40e_pf *pf)
  **/
 static int i40e_config_rss(struct i40e_pf *pf)
 {
-	const u64 default_hena =
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_SCTP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) |
-			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_SCTP) |
-			((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_OTHER) |
-			((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6) |
-			((u64)1 << I40E_FILTER_PCTYPE_L2_PAYLOAD);
-
 	/* Set of random keys generated using kernel random number generator */
 	static const u32 seed[I40E_PFQF_HKEY_MAX_INDEX + 1] = {0x41b01687,
 				0x183cfd8c, 0xce880440, 0x580cbc3c, 0x35897377,
@@ -5686,7 +5667,7 @@ static int i40e_config_rss(struct i40e_pf *pf)
 	/* By default we enable TCP/UDP with IPv4/IPv6 ptypes */
 	hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) |
 		((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32);
-	hena |= default_hena;
+	hena |= I40E_DEFAULT_RSS_HENA;
 	wr32(hw, I40E_PFQF_HENA(0), (u32)hena);
 	wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index faabf22..2992830 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -66,6 +66,26 @@ enum i40e_dyn_idx_t {
 #define I40E_TX_ITR    I40E_IDX_ITR1
 #define I40E_PE_ITR    I40E_IDX_ITR2
 
+/* Supported RSS offloads */
+#define I40E_DEFAULT_RSS_HENA ( \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_SCTP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_TCP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) | \
+	((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV4) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_TCP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_SCTP) | \
+	((u64)1 << I40E_FILTER_PCTYPE_NONF_IPV6_OTHER) | \
+	((u64)1 << I40E_FILTER_PCTYPE_FRAG_IPV6) | \
+	((u64)1 << I40E_FILTER_PCTYPE_L2_PAYLOAD))
+
 /* Supported Rx Buffer Sizes */
 #define I40E_RXBUFFER_512   512    /* Used for packet split */
 #define I40E_RXBUFFER_2048  2048
-- 
1.8.3.1

[net-next v3 15/16] i40e: update led set args

[Jeff Kirsher]

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

Add an argument to led function and refactor code to flash LED lights
correctly.

Change-Id: I00b21607ced53aaa057159503875708871946259
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_common.c    | 92 +++++++++++++++++-------
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c   |  6 +-
 drivers/net/ethernet/intel/i40e/i40e_prototype.h |  2 +-
 3 files changed, 69 insertions(+), 31 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index ca58263..e1a5a57 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -401,6 +401,38 @@ void i40e_clear_pxe_mode(struct i40e_hw *hw)
 }
 
 /**
+ * i40e_led_is_mine - helper to find matching led
+ * @hw: pointer to the hw struct
+ * @idx: index into GPIO registers
+ *
+ * returns: 0 if no match, otherwise the value of the GPIO_CTL register
+ */
+static u32 i40e_led_is_mine(struct i40e_hw *hw, int idx)
+{
+	u32 gpio_val = 0;
+	u32 port;
+
+	if (!hw->func_caps.led[idx])
+		return 0;
+
+	gpio_val = rd32(hw, I40E_GLGEN_GPIO_CTL(idx));
+	port = (gpio_val & I40E_GLGEN_GPIO_CTL_PRT_NUM_MASK) >>
+		I40E_GLGEN_GPIO_CTL_PRT_NUM_SHIFT;
+
+	/* if PRT_NUM_NA is 1 then this LED is not port specific, OR
+	 * if it is not our port then ignore
+	 */
+	if ((gpio_val & I40E_GLGEN_GPIO_CTL_PRT_NUM_NA_MASK) ||
+	    (port != hw->port))
+		return 0;
+
+	return gpio_val;
+}
+
+#define I40E_LED0 22
+#define I40E_LINK_ACTIVITY 0xC
+
+/**
  * i40e_led_get - return current on/off mode
  * @hw: pointer to the hw struct
  *
@@ -411,24 +443,20 @@ void i40e_clear_pxe_mode(struct i40e_hw *hw)
  **/
 u32 i40e_led_get(struct i40e_hw *hw)
 {
-	u32 gpio_val = 0;
 	u32 mode = 0;
-	u32 port;
 	int i;
 
-	for (i = 0; i < I40E_HW_CAP_MAX_GPIO; i++) {
-		if (!hw->func_caps.led[i])
-			continue;
-
-		gpio_val = rd32(hw, I40E_GLGEN_GPIO_CTL(i));
-		port = (gpio_val & I40E_GLGEN_GPIO_CTL_PRT_NUM_MASK)
-			>> I40E_GLGEN_GPIO_CTL_PRT_NUM_SHIFT;
+	/* as per the documentation GPIO 22-29 are the LED
+	 * GPIO pins named LED0..LED7
+	 */
+	for (i = I40E_LED0; i <= I40E_GLGEN_GPIO_CTL_MAX_INDEX; i++) {
+		u32 gpio_val = i40e_led_is_mine(hw, i);
 
-		if (port != hw->port)
+		if (!gpio_val)
 			continue;
 
-		mode = (gpio_val & I40E_GLGEN_GPIO_CTL_LED_MODE_MASK)
-				>> I40E_GLGEN_GPIO_CTL_INT_MODE_SHIFT;
+		mode = (gpio_val & I40E_GLGEN_GPIO_CTL_LED_MODE_MASK) >>
+			I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT;
 		break;
 	}
 
@@ -438,31 +466,41 @@ u32 i40e_led_get(struct i40e_hw *hw)
 /**
  * i40e_led_set - set new on/off mode
  * @hw: pointer to the hw struct
- * @mode: 0=off, else on (see EAS for mode details)
+ * @mode: 0=off, 0xf=on (else see manual for mode details)
+ * @blink: true if the LED should blink when on, false if steady
+ *
+ * if this function is used to turn on the blink it should
+ * be used to disable the blink when restoring the original state.
  **/
-void i40e_led_set(struct i40e_hw *hw, u32 mode)
+void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink)
 {
-	u32 gpio_val = 0;
-	u32 led_mode = 0;
-	u32 port;
 	int i;
 
-	for (i = 0; i < I40E_HW_CAP_MAX_GPIO; i++) {
-		if (!hw->func_caps.led[i])
-			continue;
+	if (mode & 0xfffffff0)
+		hw_dbg(hw, "invalid mode passed in %X\n", mode);
 
-		gpio_val = rd32(hw, I40E_GLGEN_GPIO_CTL(i));
-		port = (gpio_val & I40E_GLGEN_GPIO_CTL_PRT_NUM_MASK)
-			>> I40E_GLGEN_GPIO_CTL_PRT_NUM_SHIFT;
+	/* as per the documentation GPIO 22-29 are the LED
+	 * GPIO pins named LED0..LED7
+	 */
+	for (i = I40E_LED0; i <= I40E_GLGEN_GPIO_CTL_MAX_INDEX; i++) {
+		u32 gpio_val = i40e_led_is_mine(hw, i);
 
-		if (port != hw->port)
+		if (!gpio_val)
 			continue;
 
-		led_mode = (mode << I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT) &
-			    I40E_GLGEN_GPIO_CTL_LED_MODE_MASK;
 		gpio_val &= ~I40E_GLGEN_GPIO_CTL_LED_MODE_MASK;
-		gpio_val |= led_mode;
+		/* this & is a bit of paranoia, but serves as a range check */
+		gpio_val |= ((mode << I40E_GLGEN_GPIO_CTL_LED_MODE_SHIFT) &
+			     I40E_GLGEN_GPIO_CTL_LED_MODE_MASK);
+
+		if (mode == I40E_LINK_ACTIVITY)
+			blink = false;
+
+		gpio_val |= (blink ? 1 : 0) <<
+			    I40E_GLGEN_GPIO_CTL_LED_BLINK_SHIFT;
+
 		wr32(hw, I40E_GLGEN_GPIO_CTL(i), gpio_val);
+		break;
 	}
 }
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 255a138..0359c36 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -916,13 +916,13 @@ static int i40e_set_phys_id(struct net_device *netdev,
 		pf->led_status = i40e_led_get(hw);
 		return blink_freq;
 	case ETHTOOL_ID_ON:
-		i40e_led_set(hw, 0xF);
+		i40e_led_set(hw, 0xF, false);
 		break;
 	case ETHTOOL_ID_OFF:
-		i40e_led_set(hw, 0x0);
+		i40e_led_set(hw, 0x0, false);
 		break;
 	case ETHTOOL_ID_INACTIVE:
-		i40e_led_set(hw, pf->led_status);
+		i40e_led_set(hw, pf->led_status, false);
 		break;
 	}
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index 5c458bb..0b6e9fc 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -61,7 +61,7 @@ void i40e_debug_aq(struct i40e_hw *hw,
 void i40e_idle_aq(struct i40e_hw *hw);
 
 u32 i40e_led_get(struct i40e_hw *hw);
-void i40e_led_set(struct i40e_hw *hw, u32 mode);
+void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink);
 
 /* admin send queue commands */
 
-- 
1.8.3.1

[net-next v3 16/16] i40e: report VF MAC addresses correctly

[Jeff Kirsher]

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

If the user does not assign a VF MAC address, then just report it as
zero.  Attempting to guess the correct primary MAC address of the VF is a
futile and heartbreaking endeavour.

Change-Id: I2673577a160afb6fc55094c890467b44e60c7584
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 1dd13f4..cc4ebf5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -2005,7 +2005,6 @@ int i40e_ndo_get_vf_config(struct net_device *netdev,
 			   int vf_id, struct ifla_vf_info *ivi)
 {
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
-	struct i40e_mac_filter *f, *ftmp;
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
 	struct i40e_vf *vf;
@@ -2029,11 +2028,7 @@ int i40e_ndo_get_vf_config(struct net_device *netdev,
 
 	ivi->vf = vf_id;
 
-	/* first entry of the list is the default ethernet address */
-	list_for_each_entry_safe(f, ftmp, &vsi->mac_filter_list, list) {
-		memcpy(&ivi->mac, f->macaddr, I40E_ETH_LENGTH_OF_ADDRESS);
-		break;
-	}
+	memcpy(&ivi->mac, vf->default_lan_addr.addr, ETH_ALEN);
 
 	ivi->tx_rate = 0;
 	ivi->vlan = le16_to_cpu(vsi->info.pvid) & I40E_VLAN_MASK;
-- 
1.8.3.1

Re: [net-next v3 14/16] i40e: make a define from a large constant

[Joe Perches]

On Fri, 2013-12-20 at 10:21 -0800, Jeff Kirsher wrote:
> Make a define used in the header file by both VF and PF drivers.
[]
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
[]
> @@ -66,6 +66,26 @@ enum i40e_dyn_idx_t {
[]
> +/* Supported RSS offloads */
> +#define I40E_DEFAULT_RSS_HENA ( \
> +	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \

BIT_ULL

> +	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \

etc...

Re: [net-next v3 14/16] i40e: make a define from a large constant

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 807 bytes --]

On Fri, 2013-12-20 at 10:40 -0800, Joe Perches wrote:
> On Fri, 2013-12-20 at 10:21 -0800, Jeff Kirsher wrote:
> > Make a define used in the header file by both VF and PF drivers.
> []
> > diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
> []
> > @@ -66,6 +66,26 @@ enum i40e_dyn_idx_t {
> []
> > +/* Supported RSS offloads */
> > +#define I40E_DEFAULT_RSS_HENA ( \
> > +	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \
> 
> BIT_ULL
> 
> > +	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \
> 
> etc...
> 

Joe, I am not seeing widespread use of the macro BIT_ULL, so what is the
driving factor to use the macro?  I know you mentioned this on a
previous patch I pushed and wanted to ask then, but simply forgot.

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

Re: [net-next v3 07/16] i40e: Rx checksum offload for VXLAN

[Or Gerlitz]

On Fri, Dec 20, 2013 at 8:21 PM, Jeff Kirsher
<jeffrey.t.kirsher@intel.com> wrote:
> From: Joseph Gasparakis <joseph.gasparakis@intel.com>
>
> This implements receive offload for VXLAN for i40e.  The hardware
> supports checksum offload/verification of the inner/outer header.
>
> Change-Id: I450db300af6713f2044fef1191a0d1d294c13369
> Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com>
> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
> Tested-by: Kavindya Deegala <kavindya.s.deegala@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 |  6 +--
>  drivers/net/ethernet/intel/i40e/i40e_txrx.c | 57 ++++++++++++++++++++++++++---
>  drivers/net/ethernet/intel/i40e/i40e_type.h | 51 ++++++++++++++------------
>  4 files changed, 82 insertions(+), 33 deletions(-)
>
> diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
> index 5d88cf4..016b83d 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e.h
> +++ b/drivers/net/ethernet/intel/i40e/i40e.h
> @@ -29,6 +29,7 @@
>  #define _I40E_H_
>
>  #include <net/tcp.h>
> +#include <net/udp.h>
>  #include <linux/init.h>
>  #include <linux/types.h>
>  #include <linux/errno.h>
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
> index 7b759a6..a184074 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_main.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
> @@ -27,9 +27,7 @@
>
>  /* Local includes */
>  #include "i40e.h"
> -#if IS_ENABLED(CONFIG_VXLAN)
> -#include "net/vxlan.h"
> -#endif
> +#include <net/vxlan.h>
>
>  const char i40e_driver_name[] = "i40e";
>  static const char i40e_driver_string[] =
> @@ -3996,9 +3994,7 @@ static int i40e_open(struct net_device *netdev)
>                                     "couldn't set broadcast err %d aq_err %d\n",
>                                     err, pf->hw.aq.asq_last_status);
>         }
> -#if IS_ENABLED(CONFIG_VXLAN)
>         vxlan_get_rx_port(netdev);
> -#endif

the wrapping with #if IS_ENABLED(CONFIG_VXLAN) here and above is added
in patch #6 and removed in #7, I wasn't sure to follow what's the
rational




>
>         return 0;
>
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> index 01d0334..a978451 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
> @@ -860,12 +860,25 @@ static void i40e_receive_skb(struct i40e_ring *rx_ring,
>   * @skb: skb currently being received and modified
>   * @rx_status: status value of last descriptor in packet
>   * @rx_error: error value of last descriptor in packet
> + * @rx_ptype: ptype value of last descriptor in packet
>   **/
>  static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
>                                     struct sk_buff *skb,
>                                     u32 rx_status,
> -                                   u32 rx_error)
> +                                   u32 rx_error,
> +                                   u16 rx_ptype)
>  {
> +       bool ipv4_tunnel, ipv6_tunnel;
> +       __wsum rx_udp_csum;
> +       __sum16 csum;
> +       struct iphdr *iph;
> +
> +       ipv4_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT4_MAC_PAY3) &&
> +                     (rx_ptype < I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4);
> +       ipv6_tunnel = (rx_ptype > I40E_RX_PTYPE_GRENAT6_MAC_PAY3) &&
> +                     (rx_ptype < I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4);
> +
> +       skb->encapsulation = ipv4_tunnel || ipv6_tunnel;
>         skb->ip_summed = CHECKSUM_NONE;
>
>         /* Rx csum enabled and ip headers found? */
> @@ -873,13 +886,43 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
>               rx_status & (1 << I40E_RX_DESC_STATUS_L3L4P_SHIFT)))
>                 return;
>
> -       /* IP or L4 checksum error */
> +       /* 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_L4E_SHIFT) |
> +                       (1 << I40E_RX_DESC_ERROR_EIPE_SHIFT))) {
>                 vsi->back->hw_csum_rx_error++;
>                 return;
>         }
>
> +       if (ipv4_tunnel &&
> +           !(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);
> +
> +               /* Add 4 bytes for VLAN tagged packets */
> +               skb->transport_header += (skb->protocol == htons(ETH_P_8021Q) ||
> +                                         skb->protocol == htons(ETH_P_8021AD))
> +                                         ? VLAN_HLEN : 0;
> +
> +               rx_udp_csum = udp_csum(skb);
> +               iph = ip_hdr(skb);
> +               csum = csum_tcpudp_magic(
> +                               iph->saddr, iph->daddr,
> +                               (skb->len - skb_transport_offset(skb)),
> +                               IPPROTO_UDP, rx_udp_csum);
> +
> +               if (udp_hdr(skb)->check != csum) {
> +                       vsi->back->hw_csum_rx_error++;
> +                       return;
> +               }
> +       }
> +
>         skb->ip_summed = CHECKSUM_UNNECESSARY;
>  }
>
> @@ -920,6 +963,7 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
>         union i40e_rx_desc *rx_desc;
>         u32 rx_error, rx_status;
>         u64 qword;
> +       u16 rx_ptype;
>
>         rx_desc = I40E_RX_DESC(rx_ring, i);
>         qword = le64_to_cpu(rx_desc->wb.qword1.status_error_len);
> @@ -952,6 +996,8 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
>                 rx_hbo = rx_error & (1 << I40E_RX_DESC_ERROR_HBO_SHIFT);
>                 rx_error &= ~(1 << I40E_RX_DESC_ERROR_HBO_SHIFT);
>
> +               rx_ptype = (qword & I40E_RXD_QW1_PTYPE_MASK) >>
> +                          I40E_RXD_QW1_PTYPE_SHIFT;
>                 rx_bi->skb = NULL;
>
>                 /* This memory barrier is needed to keep us from reading
> @@ -1032,13 +1078,14 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)
>                 }
>
>                 skb->rxhash = i40e_rx_hash(rx_ring, rx_desc);
> -               i40e_rx_checksum(vsi, skb, rx_status, rx_error);
> -
>                 /* probably a little skewed due to removing CRC */
>                 total_rx_bytes += skb->len;
>                 total_rx_packets++;
>
>                 skb->protocol = eth_type_trans(skb, rx_ring->netdev);
> +
> +               i40e_rx_checksum(vsi, skb, rx_status, rx_error, rx_ptype);
> +
>                 vlan_tag = rx_status & (1 << I40E_RX_DESC_STATUS_L2TAG1P_SHIFT)
>                          ? le16_to_cpu(rx_desc->wb.qword0.lo_dword.l2tag1)
>                          : 0;
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
> index bb960cd..6e5f35e7 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_type.h
> +++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
> @@ -522,7 +522,8 @@ enum i40e_rx_desc_status_bits {
>         I40E_RX_DESC_STATUS_UMBCAST_SHIFT       = 9, /* 2 BITS */
>         I40E_RX_DESC_STATUS_FLM_SHIFT           = 11,
>         I40E_RX_DESC_STATUS_FLTSTAT_SHIFT       = 12, /* 2 BITS */
> -       I40E_RX_DESC_STATUS_LPBK_SHIFT          = 14
> +       I40E_RX_DESC_STATUS_LPBK_SHIFT          = 14,
> +       I40E_RX_DESC_STATUS_UDP_0_SHIFT         = 16
>  };
>
>  #define I40E_RXD_QW1_STATUS_TSYNINDX_SHIFT   I40E_RX_DESC_STATUS_TSYNINDX_SHIFT
> @@ -568,28 +569,32 @@ enum i40e_rx_desc_error_l3l4e_fcoe_masks {
>
>  /* Packet type non-ip values */
>  enum i40e_rx_l2_ptype {
> -       I40E_RX_PTYPE_L2_RESERVED               = 0,
> -       I40E_RX_PTYPE_L2_MAC_PAY2               = 1,
> -       I40E_RX_PTYPE_L2_TIMESYNC_PAY2          = 2,
> -       I40E_RX_PTYPE_L2_FIP_PAY2               = 3,
> -       I40E_RX_PTYPE_L2_OUI_PAY2               = 4,
> -       I40E_RX_PTYPE_L2_MACCNTRL_PAY2          = 5,
> -       I40E_RX_PTYPE_L2_LLDP_PAY2              = 6,
> -       I40E_RX_PTYPE_L2_ECP_PAY2               = 7,
> -       I40E_RX_PTYPE_L2_EVB_PAY2               = 8,
> -       I40E_RX_PTYPE_L2_QCN_PAY2               = 9,
> -       I40E_RX_PTYPE_L2_EAPOL_PAY2             = 10,
> -       I40E_RX_PTYPE_L2_ARP                    = 11,
> -       I40E_RX_PTYPE_L2_FCOE_PAY3              = 12,
> -       I40E_RX_PTYPE_L2_FCOE_FCDATA_PAY3       = 13,
> -       I40E_RX_PTYPE_L2_FCOE_FCRDY_PAY3        = 14,
> -       I40E_RX_PTYPE_L2_FCOE_FCRSP_PAY3        = 15,
> -       I40E_RX_PTYPE_L2_FCOE_FCOTHER_PA        = 16,
> -       I40E_RX_PTYPE_L2_FCOE_VFT_PAY3          = 17,
> -       I40E_RX_PTYPE_L2_FCOE_VFT_FCDATA        = 18,
> -       I40E_RX_PTYPE_L2_FCOE_VFT_FCRDY         = 19,
> -       I40E_RX_PTYPE_L2_FCOE_VFT_FCRSP         = 20,
> -       I40E_RX_PTYPE_L2_FCOE_VFT_FCOTHER       = 21
> +       I40E_RX_PTYPE_L2_RESERVED                       = 0,
> +       I40E_RX_PTYPE_L2_MAC_PAY2                       = 1,
> +       I40E_RX_PTYPE_L2_TIMESYNC_PAY2                  = 2,
> +       I40E_RX_PTYPE_L2_FIP_PAY2                       = 3,
> +       I40E_RX_PTYPE_L2_OUI_PAY2                       = 4,
> +       I40E_RX_PTYPE_L2_MACCNTRL_PAY2                  = 5,
> +       I40E_RX_PTYPE_L2_LLDP_PAY2                      = 6,
> +       I40E_RX_PTYPE_L2_ECP_PAY2                       = 7,
> +       I40E_RX_PTYPE_L2_EVB_PAY2                       = 8,
> +       I40E_RX_PTYPE_L2_QCN_PAY2                       = 9,
> +       I40E_RX_PTYPE_L2_EAPOL_PAY2                     = 10,
> +       I40E_RX_PTYPE_L2_ARP                            = 11,
> +       I40E_RX_PTYPE_L2_FCOE_PAY3                      = 12,
> +       I40E_RX_PTYPE_L2_FCOE_FCDATA_PAY3               = 13,
> +       I40E_RX_PTYPE_L2_FCOE_FCRDY_PAY3                = 14,
> +       I40E_RX_PTYPE_L2_FCOE_FCRSP_PAY3                = 15,
> +       I40E_RX_PTYPE_L2_FCOE_FCOTHER_PA                = 16,
> +       I40E_RX_PTYPE_L2_FCOE_VFT_PAY3                  = 17,
> +       I40E_RX_PTYPE_L2_FCOE_VFT_FCDATA                = 18,
> +       I40E_RX_PTYPE_L2_FCOE_VFT_FCRDY                 = 19,
> +       I40E_RX_PTYPE_L2_FCOE_VFT_FCRSP                 = 20,
> +       I40E_RX_PTYPE_L2_FCOE_VFT_FCOTHER               = 21,
> +       I40E_RX_PTYPE_GRENAT4_MAC_PAY3                  = 58,
> +       I40E_RX_PTYPE_GRENAT4_MACVLAN_IPV6_ICMP_PAY4    = 87,
> +       I40E_RX_PTYPE_GRENAT6_MAC_PAY3                  = 124,
> +       I40E_RX_PTYPE_GRENAT6_MACVLAN_IPV6_ICMP_PAY4    = 153
>  };
>
>  struct i40e_rx_ptype_decoded {
> --
> 1.8.3.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [net-next v3 06/16] i40e: Implementation of vxlan ndo's

[Or Gerlitz]

On Fri, Dec 20, 2013 at 8:21 PM, Jeff Kirsher
<jeffrey.t.kirsher@intel.com> wrote:
> From: Joseph Gasparakis <joseph.gasparakis@intel.com>
>
> This adds the implementation for the vxlan ndo's.  This allows
> the hardware to do RX checksum offload for inner packets on the UDP
> ports that vxlan notifies us about.
>
> Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com>
> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
> Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> ---
>  drivers/net/ethernet/intel/Kconfig               |  1 +
>  drivers/net/ethernet/intel/i40e/i40e.h           |  2 +
>  drivers/net/ethernet/intel/i40e/i40e_common.c    | 57 +++++++++++++++
>  drivers/net/ethernet/intel/i40e/i40e_main.c      | 92 ++++++++++++++++++++++++
>  drivers/net/ethernet/intel/i40e/i40e_prototype.h |  6 ++
>  drivers/net/ethernet/intel/i40e/i40e_type.h      |  1 +
>  6 files changed, 159 insertions(+)
>
> diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig
> index 41a2fa2..38e47ed 100644
> --- a/drivers/net/ethernet/intel/Kconfig
> +++ b/drivers/net/ethernet/intel/Kconfig
> @@ -244,6 +244,7 @@ config IXGBEVF
>  config I40E
>         tristate "Intel(R) Ethernet Controller XL710 Family support"
>         depends on PCI
> +       depends on VXLAN || !VXLAN
>         ---help---
>           This driver supports Intel(R) Ethernet Controller XL710 Family of
>           devices.  For more information on how to identify your adapter, go
> diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
> index 31dd265..5d88cf4 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e.h
> +++ b/drivers/net/ethernet/intel/i40e/i40e.h
> @@ -207,6 +207,8 @@ struct i40e_pf {
>         u8 atr_sample_rate;
>         bool wol_en;
>
> +       __be16  vxlan_ports[I40E_MAX_PF_UDP_OFFLOAD_PORTS];
> +
>         enum i40e_interrupt_policy int_policy;
>         u16 rx_itr_default;
>         u16 tx_itr_default;
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
> index cba1eb4..ca58263 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_common.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
> @@ -1671,6 +1671,63 @@ i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
>  }
>
>  /**
> + * i40e_aq_add_udp_tunnel
> + * @hw: pointer to the hw struct
> + * @udp_port: the UDP port to add
> + * @header_len: length of the tunneling header length in DWords
> + * @protocol_index: protocol index type
> + * @cmd_details: pointer to command details structure or NULL
> + **/
> +i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
> +                               u16 udp_port, u8 header_len,
> +                               u8 protocol_index, u8 *filter_index,
> +                               struct i40e_asq_cmd_details *cmd_details)
> +{
> +       struct i40e_aq_desc desc;
> +       struct i40e_aqc_add_udp_tunnel *cmd =
> +               (struct i40e_aqc_add_udp_tunnel *)&desc.params.raw;
> +       struct i40e_aqc_del_udp_tunnel_completion *resp =
> +               (struct i40e_aqc_del_udp_tunnel_completion *)&desc.params.raw;
> +       i40e_status status;
> +
> +       i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_udp_tunnel);
> +
> +       cmd->udp_port = cpu_to_le16(udp_port);
> +       cmd->header_len = header_len;
> +       cmd->protocol_index = protocol_index;
> +
> +       status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);

so i40e_asq_send_command is called with a spinlock held or under
rcu_read_lock from the vxlan driver code but might sleep, as it takes
a mutex, seems problematic.

> +
> +       if (!status)
> +               *filter_index = resp->index;
> +
> +       return status;
> +}
> +
> +/**
> + * i40e_aq_del_udp_tunnel
> + * @hw: pointer to the hw struct
> + * @index: filter index
> + * @cmd_details: pointer to command details structure or NULL
> + **/
> +i40e_status i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
> +                               struct i40e_asq_cmd_details *cmd_details)
> +{
> +       struct i40e_aq_desc desc;
> +       struct i40e_aqc_remove_udp_tunnel *cmd =
> +               (struct i40e_aqc_remove_udp_tunnel *)&desc.params.raw;
> +       i40e_status status;
> +
> +       i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_del_udp_tunnel);
> +
> +       cmd->index = index;
> +
> +       status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
> +
> +       return status;
> +}
> +
> +/**
>   * i40e_aq_delete_element - Delete switch element
>   * @hw: pointer to the hw struct
>   * @seid: the SEID to delete from the switch
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
> index b0cfb4c..7b759a6 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_main.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
> @@ -27,6 +27,9 @@
>
>  /* Local includes */
>  #include "i40e.h"
> +#if IS_ENABLED(CONFIG_VXLAN)
> +#include "net/vxlan.h"
> +#endif
>
>  const char i40e_driver_name[] = "i40e";
>  static const char i40e_driver_string[] =
> @@ -3993,6 +3996,9 @@ static int i40e_open(struct net_device *netdev)
>                                     "couldn't set broadcast err %d aq_err %d\n",
>                                     err, pf->hw.aq.asq_last_status);
>         }
> +#if IS_ENABLED(CONFIG_VXLAN)
> +       vxlan_get_rx_port(netdev);
> +#endif
>
>         return 0;
>
> @@ -5900,6 +5906,90 @@ static int i40e_set_features(struct net_device *netdev,
>         return 0;
>  }
>
> +/**
> + * i40e_get_vxlan_port_idx - Lookup a possibly offloaded for Rx UDP port
> + * @pf: board private structure
> + * @port: The UDP port to look up
> + *
> + * Returns the index number or I40E_MAX_PF_UDP_OFFLOAD_PORTS if port not found
> + **/
> +static u8 i40e_get_vxlan_port_idx(struct i40e_pf *pf, __be16 port)
> +{
> +       u8 i;
> +
> +       for (i = 0; i < I40E_MAX_PF_UDP_OFFLOAD_PORTS; i++) {
> +               if (pf->vxlan_ports[i] == port)
> +                       return i;
> +       }
> +
> +       return i;
> +}
> +
> +/**
> + * i40e_add_vxlan_port - Get notifications about VXLAN ports that come up
> + * @netdev: This physical port's netdev
> + * @sa_family: Socket Family that VXLAN is notifying us about
> + * @port: New UDP port number that VXLAN started listening to
> + **/
> +static void i40e_add_vxlan_port(struct net_device *netdev,
> +                                sa_family_t sa_family, __be16 port)
> +{
> +       struct i40e_netdev_priv *np = netdev_priv(netdev);
> +       struct i40e_vsi *vsi = np->vsi;
> +       struct i40e_pf *pf = vsi->back;
> +       const int vxlan_hdr_qwords = 4;
> +       struct i40e_hw *hw = &pf->hw;
> +       i40e_status aq_ret;
> +       u8 filter_index;
> +
> +       if (sa_family == AF_INET6)
> +               return;
> +
> +       aq_ret = i40e_aq_add_udp_tunnel(hw, ntohs(port), vxlan_hdr_qwords,
> +                                    I40E_AQC_TUNNEL_TYPE_UDP, &filter_index,
> +                                    NULL);
> +
> +       if (!aq_ret && filter_index < I40E_MAX_PF_UDP_OFFLOAD_PORTS)
> +               pf->vxlan_ports[filter_index] = port;
> +       else
> +               dev_warn(&pf->pdev->dev, "Could not use port %d for VXLAN offload\n",
> +                        ntohs(port));
> +}
> +
> +/**
> + * i40e_del_vxlan_port - Get notifications about VXLAN ports that go away
> + * @netdev: This physical port's netdev
> + * @sa_family: Socket Family that VXLAN is notifying us about
> + * @port: UDP port number that VXLAN stopped listening to
> + **/
> +static void i40e_del_vxlan_port(struct net_device *netdev,
> +                                sa_family_t sa_family, __be16 port)
> +{
> +       struct i40e_netdev_priv *np = netdev_priv(netdev);
> +       struct i40e_vsi *vsi = np->vsi;
> +       struct i40e_pf *pf = vsi->back;
> +       struct i40e_hw *hw = &pf->hw;
> +       i40e_status aq_ret;
> +       u8 filter_index;
> +
> +       if (sa_family == AF_INET6)
> +               return;
> +
> +       filter_index = i40e_get_vxlan_port_idx(pf, port);
> +
> +       if (filter_index < I40E_MAX_PF_UDP_OFFLOAD_PORTS) {
> +               aq_ret = i40e_aq_del_udp_tunnel(hw, filter_index, NULL);
> +               if (!aq_ret)
> +                       pf->vxlan_ports[filter_index] = 0;
> +               else
> +                       dev_warn(&pf->pdev->dev, "Could not remove VXLAN port %d\n",
> +                                ntohs(port));
> +       } else {
> +               dev_warn(&pf->pdev->dev, "Received notification to remove VXLAN port %d but port was not added before.\n",
> +                        ntohs(port));
> +       }
> +}
> +
>  static const struct net_device_ops i40e_netdev_ops = {
>         .ndo_open               = i40e_open,
>         .ndo_stop               = i40e_close,
> @@ -5921,6 +6011,8 @@ static const struct net_device_ops i40e_netdev_ops = {
>         .ndo_set_vf_vlan        = i40e_ndo_set_vf_port_vlan,
>         .ndo_set_vf_tx_rate     = i40e_ndo_set_vf_bw,
>         .ndo_get_vf_config      = i40e_ndo_get_vf_config,
> +       .ndo_add_vxlan_port     = i40e_add_vxlan_port,
> +       .ndo_del_vxlan_port     = i40e_del_vxlan_port,
>  };
>
>  /**
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
> index db7bf93..5c458bb 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
> +++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
> @@ -157,6 +157,12 @@ i40e_status i40e_aq_stop_lldp(struct i40e_hw *hw, bool shutdown_agent,
>                                 struct i40e_asq_cmd_details *cmd_details);
>  i40e_status i40e_aq_start_lldp(struct i40e_hw *hw,
>                                 struct i40e_asq_cmd_details *cmd_details);
> +i40e_status i40e_aq_add_udp_tunnel(struct i40e_hw *hw,
> +                               u16 udp_port, u8 header_len,
> +                               u8 protocol_index, u8 *filter_index,
> +                               struct i40e_asq_cmd_details *cmd_details);
> +i40e_status i40e_aq_del_udp_tunnel(struct i40e_hw *hw, u8 index,
> +                               struct i40e_asq_cmd_details *cmd_details);
>  i40e_status i40e_aq_delete_element(struct i40e_hw *hw, u16 seid,
>                                 struct i40e_asq_cmd_details *cmd_details);
>  i40e_status i40e_aq_mac_address_write(struct i40e_hw *hw,
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
> index d683c30..bb960cd 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_type.h
> +++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
> @@ -59,6 +59,7 @@
>  #define I40E_MAX_VSI_QP                        16
>  #define I40E_MAX_VF_VSI                        3
>  #define I40E_MAX_CHAINED_RX_BUFFERS    5
> +#define I40E_MAX_PF_UDP_OFFLOAD_PORTS  16
>
>  /* Max default timeout in ms, */
>  #define I40E_MAX_NVM_TIMEOUT           18000
> --
> 1.8.3.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [net-next v3 06/16] i40e: Implementation of vxlan ndo's

[David Miller]

From: Or Gerlitz <or.gerlitz@gmail.com>
Date: Fri, 20 Dec 2013 23:06:06 +0200

> On Fri, Dec 20, 2013 at 8:21 PM, Jeff Kirsher
> <jeffrey.t.kirsher@intel.com> wrote:
>> @@ -244,6 +244,7 @@ config IXGBEVF
>>  config I40E
>>         tristate "Intel(R) Ethernet Controller XL710 Family support"
>>         depends on PCI
>> +       depends on VXLAN || !VXLAN

I definitely want a clear detailed explanation for this gross
dependency.

>> +       i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_udp_tunnel);
>> +
>> +       cmd->udp_port = cpu_to_le16(udp_port);
>> +       cmd->header_len = header_len;
>> +       cmd->protocol_index = protocol_index;
>> +
>> +       status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
> 
> so i40e_asq_send_command is called with a spinlock held or under
> rcu_read_lock from the vxlan driver code but might sleep, as it takes
> a mutex, seems problematic.

Indeed this looks like a bug.

Re: [net-next v3 14/16] i40e: make a define from a large constant

[Joe Perches]

On Fri, 2013-12-20 at 12:12 -0800, Jeff Kirsher wrote:
> On Fri, 2013-12-20 at 10:40 -0800, Joe Perches wrote:
> > On Fri, 2013-12-20 at 10:21 -0800, Jeff Kirsher wrote:
> > > Make a define used in the header file by both VF and PF drivers.
> > []
> > > diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
> > []
> > > @@ -66,6 +66,26 @@ enum i40e_dyn_idx_t {
> > []
> > > +/* Supported RSS offloads */
> > > +#define I40E_DEFAULT_RSS_HENA ( \
> > > +	((u64)1 << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \
> > 
> > BIT_ULL
> > 
> > > +	((u64)1 << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \
> > 
> > etc...
> > 
> 
> Joe, I am not seeing widespread use of the macro BIT_ULL, so what is the
> driving factor to use the macro?

Hey Jeff.

It was just introduced in October 2013.

commit bfd1ff6375c82930bfb3b401eee2c96720fa8e84
Author: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Date:   Fri Oct 11 16:54:59 2013 -0700

    bitops: Introduce BIT_ULL
    
    Adding BIT(x) equivalent for unsigned long long type, BIT_ULL(x). Also
    added BIT_ULL_MASK and BIT_ULL_WORD.
    

>   I know you mentioned this on a
> previous patch I pushed and wanted to ask then, but simply forgot.

Just consistency and symmetry and a few
bug reduction possibilities using standard
mechanisms.

Re: [net-next v3 06/16] i40e: Implementation of vxlan ndo's

[Joseph Gasparakis]

On Fri, 20 Dec 2013, David Miller wrote:

> From: Or Gerlitz <or.gerlitz@gmail.com>
> Date: Fri, 20 Dec 2013 23:06:06 +0200
> 
> > On Fri, Dec 20, 2013 at 8:21 PM, Jeff Kirsher
> > <jeffrey.t.kirsher@intel.com> wrote:
> >> @@ -244,6 +244,7 @@ config IXGBEVF
> >>  config I40E
> >>         tristate "Intel(R) Ethernet Controller XL710 Family support"
> >>         depends on PCI
> >> +       depends on VXLAN || !VXLAN
> 
> I definitely want a clear detailed explanation for this gross
> dependency.

Right now the code in vxlan.h is:

#if IS_ENABLED(CONFIG_VXLAN)
void vxlan_get_rx_port(struct net_device *netdev);
#else
static inline void vxlan_get_rx_port(struct net_device *netdev)
{
}
#endif

so the function can be called from drivers whether vxlan is enabled in 
.config or not. However, the "depends on VXLAN || !VXLAN" is needed when 
the driver is compiled as a built-in, otherwise we get an undefined 
reference for the above function. If this is not the right way to fix the 
dependency please let me know and I will fix it accordingly.


> 
> >> +       i40e_fill_default_direct_cmd_desc(&desc, i40e_aqc_opc_add_udp_tunnel);
> >> +
> >> +       cmd->udp_port = cpu_to_le16(udp_port);
> >> +       cmd->header_len = header_len;
> >> +       cmd->protocol_index = protocol_index;
> >> +
> >> +       status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
> > 
> > so i40e_asq_send_command is called with a spinlock held or under
> > rcu_read_lock from the vxlan driver code but might sleep, as it takes
> > a mutex, seems problematic.
> 
> Indeed this looks like a bug.

Ok, the comment makes sense, I will rework this.

> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [net-next v3 06/16] i40e: Implementation of vxlan ndo's

[David Miller]

From: Joseph Gasparakis <joseph.gasparakis@intel.com>
Date: Fri, 20 Dec 2013 15:25:44 -0800 (PST)

> Right now the code in vxlan.h is:
> 
> #if IS_ENABLED(CONFIG_VXLAN)
> void vxlan_get_rx_port(struct net_device *netdev);
> #else
> static inline void vxlan_get_rx_port(struct net_device *netdev)
> {
> }
> #endif
> 
> so the function can be called from drivers whether vxlan is enabled in 
> .config or not. However, the "depends on VXLAN || !VXLAN" is needed when 
> the driver is compiled as a built-in, otherwise we get an undefined 
> reference for the above function. If this is not the right way to fix the 
> dependency please let me know and I will fix it accordingly.

Ok, can you write this with the more canonical:

	(X || X=n)

That's how we write it out to handle the same issue with
IPV6.

Thanks.

Re: [net-next v3 06/16] i40e: Implementation of vxlan ndo's

[Joseph Gasparakis]

On Fri, 20 Dec 2013, David Miller wrote:

> From: Joseph Gasparakis <joseph.gasparakis@intel.com>
> Date: Fri, 20 Dec 2013 15:25:44 -0800 (PST)
> 
> > Right now the code in vxlan.h is:
> > 
> > #if IS_ENABLED(CONFIG_VXLAN)
> > void vxlan_get_rx_port(struct net_device *netdev);
> > #else
> > static inline void vxlan_get_rx_port(struct net_device *netdev)
> > {
> > }
> > #endif
> > 
> > so the function can be called from drivers whether vxlan is enabled in 
> > .config or not. However, the "depends on VXLAN || !VXLAN" is needed when 
> > the driver is compiled as a built-in, otherwise we get an undefined 
> > reference for the above function. If this is not the right way to fix the 
> > dependency please let me know and I will fix it accordingly.
> 
> Ok, can you write this with the more canonical:
> 
> 	(X || X=n)
> 
> That's how we write it out to handle the same issue with
> IPV6.
> 
> Thanks.

Will do. Thanks.

> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>