[net-next 00/15][pull request] 10GbE Intel Wired LAN Driver Updates 2017-04-18

[net-next 00/15][pull request] 10GbE Intel Wired LAN Driver Updates 2017-04-18

[Jeff Kirsher]

This series contains updates to mainly ixgbe with only one ixgbevf change.

Usha adds a check to ensure the creation of number of VF's is valid based
on the traffic classes configured, all to avoid transmit hangs.

Joe Perches reduces the use of pr_cont since the output can be interleaved
by other processes.

Tony cleans up the code overwriting the KX4 config, which is configured by
the NVM.  Adds a check for MMNGC.MNG_VETO, to resolve an issue where we
were getting a link loss for the BMC when loading the driver.

Don fixes up SGMII x553 config details which were missed in earlier
implementations.  Added support for x552 XFI backplane interface support.
Cleaned up an unused define, which was causing confusion on supported
devices.

Emil fixes a link issue on KR parts by making sure the default setting is
set.  Refactors the code so that the code for allocating memory for the
list of MAC addresses that the VFs can use into its own function.  Made
some code cleans to help readability and ensure notification of SRIOV
being enabled is done upon completion.  Fixed an issue where if we failed
to allocate vfinfo in __ixgbe_enable_sriov() the driver would crash with
a NULL pointer dereference.

Philippe Reynes updates ixgbevf to use the new API for
{get|set}_link_ksettings.

Alex increases the headroom allocation when using build_skb() on a
system with 4K pages.  Fixed an issue in ixgbe_dump() where we were no
longer clearing the status bit.

The following are changes since commit 4116c97689b9b1732ac5b68afd922406f9fc842e:
  Merge branch 'ftgmac100-batch5-features'
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue 10GbE

Alexander Duyck (2):
  ixgbe: Add support for maximum headroom when using build_skb
  ixgbe: Fix output from ixgbe_dump

Don Skidmore (4):
  ixgbe: Complete support for X553 sgmii
  ixgbe: Add X552 XFI backplane support
  ixgbe: list X553 backplane speeds correctly
  ixgbe: Remove unused define

Emil Tantilov (4):
  ixgbe: add default setup_link for x550em_a MAC type
  ixgbe: move num_vfs_macvlans allocation into separate function
  ixgbe: return early instead of wrap block in if statement
  ixgbe: do not use adapter->num_vfs when setting VFs via module
    parameter

Joe Perches (1):
  ixgbe: Remove pr_cont uses

Philippe Reynes (1):
  ixgbevf: use new api ethtool_{get|set}_link_ksettings

Tony Nguyen (2):
  ixgbe: Remove driver config for KX4 PHY
  ixgbe: add check for VETO bit when configuring link for KR

Usha Ketineni (1):
  ixgbe: Avoid Tx hang by not allowing more than the number of VFs
    supported.

 drivers/net/ethernet/intel/ixgbe/ixgbe.h         |  55 +++++++-
 drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c |   1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c    | 101 +++++++-------
 drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c     |   3 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c   | 169 +++++++++++++----------
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h   |   5 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_type.h    |  22 +--
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c    | 102 ++++++++------
 drivers/net/ethernet/intel/ixgbevf/ethtool.c     |  22 +--
 9 files changed, 287 insertions(+), 193 deletions(-)

-- 
2.12.2

[net-next 01/15] ixgbe: Avoid Tx hang by not allowing more than the number of VFs supported.

[Jeff Kirsher]

From: Usha Ketineni <usha.k.ketineni@intel.com>

When DCB is enabled, add checks to ensure creation of number of VF's is
valid based on the traffic classes configured by the device.

Signed-off-by: Usha Ketineni <usha.k.ketineni@intel.com>
Tested-by: Ronald Bynoe <ronald.j.bynoe@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c | 26 +++++++++++++++++++++++---
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h |  3 +++
 2 files changed, 26 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
index 044cb44747cf..39e109da9bd9 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
@@ -298,6 +298,7 @@ static int ixgbe_pci_sriov_enable(struct pci_dev *dev, int num_vfs)
 #ifdef CONFIG_PCI_IOV
 	struct ixgbe_adapter *adapter = pci_get_drvdata(dev);
 	int err = 0;
+	u8 num_tc;
 	int i;
 	int pre_existing_vfs = pci_num_vf(dev);
 
@@ -310,16 +311,35 @@ static int ixgbe_pci_sriov_enable(struct pci_dev *dev, int num_vfs)
 		return err;
 
 	/* While the SR-IOV capability structure reports total VFs to be 64,
-	 * we have to limit the actual number allocated based on two factors.
+	 * we limit the actual number allocated as below based on two factors.
+	 *    Num_TCs	MAX_VFs
+	 *	1	  63
+	 *	<=4	  31
+	 *	>4	  15
 	 * First, we reserve some transmit/receive resources for the PF.
 	 * Second, VMDQ also uses the same pools that SR-IOV does. We need to
 	 * account for this, so that we don't accidentally allocate more VFs
 	 * than we have available pools. The PCI bus driver already checks for
 	 * other values out of range.
 	 */
-	if ((num_vfs + adapter->num_rx_pools) > IXGBE_MAX_VF_FUNCTIONS)
-		return -EPERM;
+	num_tc = netdev_get_num_tc(adapter->netdev);
 
+	if (num_tc > 4) {
+		if ((num_vfs + adapter->num_rx_pools) > IXGBE_MAX_VFS_8TC) {
+			e_dev_err("Currently the device is configured with %d TCs, Creating more than %d VFs is not allowed\n", num_tc, IXGBE_MAX_VFS_8TC);
+			return -EPERM;
+		}
+	} else if ((num_tc > 1) && (num_tc <= 4)) {
+		if ((num_vfs + adapter->num_rx_pools) > IXGBE_MAX_VFS_4TC) {
+			e_dev_err("Currently the device is configured with %d TCs, Creating more than %d VFs is not allowed\n", num_tc, IXGBE_MAX_VFS_4TC);
+			return -EPERM;
+		}
+	} else {
+		if ((num_vfs + adapter->num_rx_pools) > IXGBE_MAX_VFS_1TC) {
+			e_dev_err("Currently the device is configured with %d TCs, Creating more than %d VFs is not allowed\n", num_tc, IXGBE_MAX_VFS_1TC);
+			return -EPERM;
+		}
+	}
 	adapter->num_vfs = num_vfs;
 
 	err = __ixgbe_enable_sriov(adapter);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
index 0c7977d27b71..3166fd164e51 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
@@ -33,6 +33,9 @@
  *  63 (IXGBE_MAX_VF_FUNCTIONS - 1)
  */
 #define IXGBE_MAX_VFS_DRV_LIMIT  (IXGBE_MAX_VF_FUNCTIONS - 1)
+#define IXGBE_MAX_VFS_1TC		IXGBE_MAX_VF_FUNCTIONS
+#define IXGBE_MAX_VFS_4TC		32
+#define IXGBE_MAX_VFS_8TC		16
 
 #ifdef CONFIG_PCI_IOV
 void ixgbe_restore_vf_multicasts(struct ixgbe_adapter *adapter);
-- 
2.12.2

[net-next 02/15] ixgbe: Remove pr_cont uses

[Jeff Kirsher]

From: Joe Perches <joe@perches.com>

As pr_cont output can be interleaved by other processes,
using pr_cont should be avoided where possible.

Miscellanea:

- Use a temporary pointer to hold the next descriptions and
  consolidate the pr_cont uses
- Use the temporary buffer to hold the 8 u32 register values and
  emit those in a single go
- Coalesce formats and logging neatening around those changes
- Fix a defective output for the rx ring entry description when
  also emitting rx_buffer_info data

This reduces overall object size a tiny bit too.

$ size drivers/net/ethernet/intel/ixgbe/*.o*
   text	   data	    bss	    dec	    hex	filename
  62167	    728	     12	  62907	   f5bb	drivers/net/ethernet/intel/ixgbe/ixgbe_main.o.new
  62273	    728	     12	  63013	   f625	drivers/net/ethernet/intel/ixgbe/ixgbe_main.o.old

Signed-off-by: Joe Perches <joe@perches.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 84 +++++++++++++++------------
 1 file changed, 47 insertions(+), 37 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 852a2e7e25ed..848dea922b2a 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -508,7 +508,7 @@ static const struct ixgbe_reg_info ixgbe_reg_info_tbl[] = {
  */
 static void ixgbe_regdump(struct ixgbe_hw *hw, struct ixgbe_reg_info *reginfo)
 {
-	int i = 0, j = 0;
+	int i;
 	char rname[16];
 	u32 regs[64];
 
@@ -570,17 +570,21 @@ static void ixgbe_regdump(struct ixgbe_hw *hw, struct ixgbe_reg_info *reginfo)
 			regs[i] = IXGBE_READ_REG(hw, IXGBE_TXDCTL(i));
 		break;
 	default:
-		pr_info("%-15s %08x\n", reginfo->name,
-			IXGBE_READ_REG(hw, reginfo->ofs));
+		pr_info("%-15s %08x\n",
+			reginfo->name, IXGBE_READ_REG(hw, reginfo->ofs));
 		return;
 	}
 
-	for (i = 0; i < 8; i++) {
-		snprintf(rname, 16, "%s[%d-%d]", reginfo->name, i*8, i*8+7);
-		pr_err("%-15s", rname);
+	i = 0;
+	while (i < 64) {
+		int j;
+		char buf[9 * 8 + 1];
+		char *p = buf;
+
+		snprintf(rname, 16, "%s[%d-%d]", reginfo->name, i, i + 7);
 		for (j = 0; j < 8; j++)
-			pr_cont(" %08x", regs[i*8+j]);
-		pr_cont("\n");
+			p += sprintf(p, " %08x", regs[i++]);
+		pr_err("%-15s%s\n", rname, buf);
 	}
 
 }
@@ -701,7 +705,18 @@ static void ixgbe_dump(struct ixgbe_adapter *adapter)
 			tx_buffer = &tx_ring->tx_buffer_info[i];
 			u0 = (struct my_u0 *)tx_desc;
 			if (dma_unmap_len(tx_buffer, len) > 0) {
-				pr_info("T [0x%03X]    %016llX %016llX %016llX %08X %p %016llX %p",
+				const char *ring_desc;
+
+				if (i == tx_ring->next_to_use &&
+				    i == tx_ring->next_to_clean)
+					ring_desc = " NTC/U";
+				else if (i == tx_ring->next_to_use)
+					ring_desc = " NTU";
+				else if (i == tx_ring->next_to_clean)
+					ring_desc = " NTC";
+				else
+					ring_desc = "";
+				pr_info("T [0x%03X]    %016llX %016llX %016llX %08X %p %016llX %p%s",
 					i,
 					le64_to_cpu(u0->a),
 					le64_to_cpu(u0->b),
@@ -709,16 +724,8 @@ static void ixgbe_dump(struct ixgbe_adapter *adapter)
 					dma_unmap_len(tx_buffer, len),
 					tx_buffer->next_to_watch,
 					(u64)tx_buffer->time_stamp,
-					tx_buffer->skb);
-				if (i == tx_ring->next_to_use &&
-					i == tx_ring->next_to_clean)
-					pr_cont(" NTC/U\n");
-				else if (i == tx_ring->next_to_use)
-					pr_cont(" NTU\n");
-				else if (i == tx_ring->next_to_clean)
-					pr_cont(" NTC\n");
-				else
-					pr_cont("\n");
+					tx_buffer->skb,
+					ring_desc);
 
 				if (netif_msg_pktdata(adapter) &&
 				    tx_buffer->skb)
@@ -797,34 +804,45 @@ static void ixgbe_dump(struct ixgbe_adapter *adapter)
 		pr_info("------------------------------------\n");
 		pr_info("RX QUEUE INDEX = %d\n", rx_ring->queue_index);
 		pr_info("------------------------------------\n");
-		pr_info("%s%s%s",
+		pr_info("%s%s%s\n",
 			"R  [desc]      [ PktBuf     A0] ",
 			"[  HeadBuf   DD] [bi->dma       ] [bi->skb       ] ",
-			"<-- Adv Rx Read format\n");
-		pr_info("%s%s%s",
+			"<-- Adv Rx Read format");
+		pr_info("%s%s%s\n",
 			"RWB[desc]      [PcsmIpSHl PtRs] ",
 			"[vl er S cks ln] ---------------- [bi->skb       ] ",
-			"<-- Adv Rx Write-Back format\n");
+			"<-- Adv Rx Write-Back format");
 
 		for (i = 0; i < rx_ring->count; i++) {
+			const char *ring_desc;
+
+			if (i == rx_ring->next_to_use)
+				ring_desc = " NTU";
+			else if (i == rx_ring->next_to_clean)
+				ring_desc = " NTC";
+			else
+				ring_desc = "";
+
 			rx_buffer_info = &rx_ring->rx_buffer_info[i];
 			rx_desc = IXGBE_RX_DESC(rx_ring, i);
 			u0 = (struct my_u0 *)rx_desc;
 			staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
 			if (staterr & IXGBE_RXD_STAT_DD) {
 				/* Descriptor Done */
-				pr_info("RWB[0x%03X]     %016llX "
-					"%016llX ---------------- %p", i,
+				pr_info("RWB[0x%03X]     %016llX %016llX ---------------- %p%s\n",
+					i,
 					le64_to_cpu(u0->a),
 					le64_to_cpu(u0->b),
-					rx_buffer_info->skb);
+					rx_buffer_info->skb,
+					ring_desc);
 			} else {
-				pr_info("R  [0x%03X]     %016llX "
-					"%016llX %016llX %p", i,
+				pr_info("R  [0x%03X]     %016llX %016llX %016llX %p%s\n",
+					i,
 					le64_to_cpu(u0->a),
 					le64_to_cpu(u0->b),
 					(u64)rx_buffer_info->dma,
-					rx_buffer_info->skb);
+					rx_buffer_info->skb,
+					ring_desc);
 
 				if (netif_msg_pktdata(adapter) &&
 				    rx_buffer_info->dma) {
@@ -835,14 +853,6 @@ static void ixgbe_dump(struct ixgbe_adapter *adapter)
 					   ixgbe_rx_bufsz(rx_ring), true);
 				}
 			}
-
-			if (i == rx_ring->next_to_use)
-				pr_cont(" NTU\n");
-			else if (i == rx_ring->next_to_clean)
-				pr_cont(" NTC\n");
-			else
-				pr_cont("\n");
-
 		}
 	}
 }
-- 
2.12.2

[net-next 03/15] ixgbe: Remove driver config for KX4 PHY

[Jeff Kirsher]

From: Tony Nguyen <anthony.l.nguyen@intel.com>

The KX4 PHY is configured by the NVM.  Currently, the driver is overwriting
the config; remove the code associated with KX4 configuration.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Krishneil Singh  <krishneil.k.singh@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c  |  3 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_type.h | 12 --------
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 40 +--------------------------
 3 files changed, 3 insertions(+), 52 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
index e55b2602f371..654a402f0e9e 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
@@ -792,7 +792,8 @@ s32 ixgbe_setup_phy_link_speed_generic(struct ixgbe_hw *hw,
 		hw->phy.autoneg_advertised |= IXGBE_LINK_SPEED_10_FULL;
 
 	/* Setup link based on the new speed settings */
-	hw->phy.ops.setup_link(hw);
+	if (hw->phy.ops.setup_link)
+		hw->phy.ops.setup_link(hw);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
index 1d07f2ead914..c0f6b9503439 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
@@ -3754,15 +3754,6 @@ struct ixgbe_info {
 #define IXGBE_KRM_TX_COEFF_CTRL_1_CZERO_EN		BIT(3)
 #define IXGBE_KRM_TX_COEFF_CTRL_1_OVRRD_EN		BIT(31)
 
-#define IXGBE_KX4_LINK_CNTL_1				0x4C
-#define IXGBE_KX4_LINK_CNTL_1_TETH_AN_CAP_KX		BIT(16)
-#define IXGBE_KX4_LINK_CNTL_1_TETH_AN_CAP_KX4		BIT(17)
-#define IXGBE_KX4_LINK_CNTL_1_TETH_EEE_CAP_KX		BIT(24)
-#define IXGBE_KX4_LINK_CNTL_1_TETH_EEE_CAP_KX4		BIT(25)
-#define IXGBE_KX4_LINK_CNTL_1_TETH_AN_ENABLE		BIT(29)
-#define IXGBE_KX4_LINK_CNTL_1_TETH_FORCE_LINK_UP	BIT(30)
-#define IXGBE_KX4_LINK_CNTL_1_TETH_AN_RESTART		BIT(31)
-
 #define IXGBE_SB_IOSF_INDIRECT_CTRL		0x00011144
 #define IXGBE_SB_IOSF_INDIRECT_DATA		0x00011148
 
@@ -3779,9 +3770,6 @@ struct ixgbe_info {
 #define IXGBE_SB_IOSF_CTRL_BUSY_SHIFT		31
 #define IXGBE_SB_IOSF_CTRL_BUSY		BIT(IXGBE_SB_IOSF_CTRL_BUSY_SHIFT)
 #define IXGBE_SB_IOSF_TARGET_KR_PHY	0
-#define IXGBE_SB_IOSF_TARGET_KX4_UNIPHY	1
-#define IXGBE_SB_IOSF_TARGET_KX4_PCS0	2
-#define IXGBE_SB_IOSF_TARGET_KX4_PCS1	3
 
 #define IXGBE_NW_MNG_IF_SEL		0x00011178
 #define IXGBE_NW_MNG_IF_SEL_MDIO_ACT		BIT(1)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
index 200f847fd8f3..0a2eacff7327 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
@@ -2473,44 +2473,6 @@ static s32 ixgbe_setup_kr_speed_x550em(struct ixgbe_hw *hw,
 	return ixgbe_restart_an_internal_phy_x550em(hw);
 }
 
-/** ixgbe_setup_kx4_x550em - Configure the KX4 PHY.
- *  @hw: pointer to hardware structure
- *
- *   Configures the integrated KX4 PHY.
- **/
-static s32 ixgbe_setup_kx4_x550em(struct ixgbe_hw *hw)
-{
-	s32 status;
-	u32 reg_val;
-
-	status = hw->mac.ops.read_iosf_sb_reg(hw, IXGBE_KX4_LINK_CNTL_1,
-					      IXGBE_SB_IOSF_TARGET_KX4_PCS0 +
-					      hw->bus.lan_id, &reg_val);
-	if (status)
-		return status;
-
-	reg_val &= ~(IXGBE_KX4_LINK_CNTL_1_TETH_AN_CAP_KX4 |
-		     IXGBE_KX4_LINK_CNTL_1_TETH_AN_CAP_KX);
-
-	reg_val |= IXGBE_KX4_LINK_CNTL_1_TETH_AN_ENABLE;
-
-	/* Advertise 10G support. */
-	if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_10GB_FULL)
-		reg_val |= IXGBE_KX4_LINK_CNTL_1_TETH_AN_CAP_KX4;
-
-	/* Advertise 1G support. */
-	if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_1GB_FULL)
-		reg_val |= IXGBE_KX4_LINK_CNTL_1_TETH_AN_CAP_KX;
-
-	/* Restart auto-negotiation. */
-	reg_val |= IXGBE_KX4_LINK_CNTL_1_TETH_AN_RESTART;
-	status = hw->mac.ops.write_iosf_sb_reg(hw, IXGBE_KX4_LINK_CNTL_1,
-					       IXGBE_SB_IOSF_TARGET_KX4_PCS0 +
-					       hw->bus.lan_id, reg_val);
-
-	return status;
-}
-
 /**
  * ixgbe_setup_kr_x550em - Configure the KR PHY
  * @hw: pointer to hardware structure
@@ -3134,7 +3096,7 @@ static s32 ixgbe_init_phy_ops_X550em(struct ixgbe_hw *hw)
 	/* Set functions pointers based on phy type */
 	switch (hw->phy.type) {
 	case ixgbe_phy_x550em_kx4:
-		phy->ops.setup_link = ixgbe_setup_kx4_x550em;
+		phy->ops.setup_link = NULL;
 		phy->ops.read_reg = ixgbe_read_phy_reg_x550em;
 		phy->ops.write_reg = ixgbe_write_phy_reg_x550em;
 		break;
-- 
2.12.2

[net-next 04/15] ixgbe: Complete support for X553 sgmii

[Jeff Kirsher]

From: Don Skidmore <donald.c.skidmore@intel.com>

The initial patches supporting X553 sgmii forgot some details.  This patch
should cover those missing spots.

Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Tested-by: Krishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 29 +++++++++++++++++++++++++--
 1 file changed, 27 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
index 0a2eacff7327..633079eaf8fb 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
@@ -334,6 +334,16 @@ static s32 ixgbe_identify_phy_x550em(struct ixgbe_hw *hw)
 	case IXGBE_DEV_ID_X550EM_X_1G_T:
 	case IXGBE_DEV_ID_X550EM_X_10G_T:
 		return ixgbe_identify_phy_generic(hw);
+	case IXGBE_DEV_ID_X550EM_A_1G_T:
+	case IXGBE_DEV_ID_X550EM_A_1G_T_L:
+		hw->phy.type = ixgbe_phy_fw;
+		hw->phy.ops.read_reg = NULL;
+		hw->phy.ops.write_reg = NULL;
+		if (hw->bus.lan_id)
+			hw->phy.phy_semaphore_mask |= IXGBE_GSSR_PHY1_SM;
+		else
+			hw->phy.phy_semaphore_mask |= IXGBE_GSSR_PHY0_SM;
+		break;
 	default:
 		break;
 	}
@@ -2215,8 +2225,20 @@ static s32 ixgbe_get_link_capabilities_X550em(struct ixgbe_hw *hw,
 		else
 			*speed = IXGBE_LINK_SPEED_10GB_FULL;
 	} else {
-		*speed = IXGBE_LINK_SPEED_10GB_FULL |
-			 IXGBE_LINK_SPEED_1GB_FULL;
+		switch (hw->phy.type) {
+		case ixgbe_phy_x550em_kx4:
+			*speed = IXGBE_LINK_SPEED_1GB_FULL |
+				 IXGBE_LINK_SPEED_2_5GB_FULL |
+				 IXGBE_LINK_SPEED_10GB_FULL;
+			break;
+		case ixgbe_phy_sgmii:
+			*speed = IXGBE_LINK_SPEED_1GB_FULL;
+			break;
+		default:
+			*speed = IXGBE_LINK_SPEED_10GB_FULL |
+				 IXGBE_LINK_SPEED_1GB_FULL;
+			break;
+		}
 		*autoneg = true;
 	}
 	return 0;
@@ -3126,6 +3148,9 @@ static s32 ixgbe_init_phy_ops_X550em(struct ixgbe_hw *hw)
 		phy->ops.handle_lasi = ixgbe_handle_lasi_ext_t_x550em;
 		phy->ops.reset = ixgbe_reset_phy_t_X550em;
 		break;
+	case ixgbe_phy_sgmii:
+		phy->ops.setup_link = NULL;
+		break;
 	case ixgbe_phy_fw:
 		phy->ops.setup_link = ixgbe_setup_fw_link;
 		phy->ops.reset = ixgbe_reset_phy_fw;
-- 
2.12.2

[net-next 05/15] ixgbe: Add X552 XFI backplane support

[Jeff Kirsher]

From: Don Skidmore <donald.c.skidmore@intel.com>

This patch add support for X552 XFI backplane interface.  The XFI
backplane requires a custom tuned link.  HW/FW owns the link config
for XF backplane and SW must not interfere with it.

Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Tested-by: Krishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c |  1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c    |  1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_type.h    |  2 ++
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c    | 14 ++++++++++++++
 4 files changed, 18 insertions(+)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 0da0752fedef..59730ede4746 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -179,6 +179,7 @@ static u32 ixgbe_get_supported_10gtypes(struct ixgbe_hw *hw)
 	case IXGBE_DEV_ID_82598_BX:
 	case IXGBE_DEV_ID_82599_KR:
 	case IXGBE_DEV_ID_X550EM_X_KR:
+	case IXGBE_DEV_ID_X550EM_X_XFI:
 		return SUPPORTED_10000baseKR_Full;
 	default:
 		return SUPPORTED_10000baseKX4_Full |
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 848dea922b2a..e250c1d54a11 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -131,6 +131,7 @@ static const struct pci_device_id ixgbe_pci_tbl[] = {
 	{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550T), board_X550},
 	{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550T1), board_X550},
 	{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_X_KX4), board_X550EM_x},
+	{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_X_XFI), board_X550EM_x},
 	{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_X_KR), board_X550EM_x},
 	{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_X_10G_T), board_X550EM_x},
 	{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X550EM_X_SFP), board_X550EM_x},
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
index c0f6b9503439..22869aaec3d4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
@@ -85,6 +85,7 @@
 #define IXGBE_DEV_ID_X550EM_X_SFP	0x15AC
 #define IXGBE_DEV_ID_X550EM_X_10G_T	0x15AD
 #define IXGBE_DEV_ID_X550EM_X_1G_T	0x15AE
+#define IXGBE_DEV_ID_X550EM_X_XFI	0x15B0
 #define IXGBE_DEV_ID_X550EM_A_KR	0x15C2
 #define IXGBE_DEV_ID_X550EM_A_KR_L	0x15C3
 #define IXGBE_DEV_ID_X550EM_A_SFP_N	0x15C4
@@ -3128,6 +3129,7 @@ enum ixgbe_phy_type {
 	ixgbe_phy_aq,
 	ixgbe_phy_x550em_kr,
 	ixgbe_phy_x550em_kx4,
+	ixgbe_phy_x550em_xfi,
 	ixgbe_phy_x550em_ext_t,
 	ixgbe_phy_cu_unknown,
 	ixgbe_phy_qt,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
index 633079eaf8fb..5127e5e759f2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
@@ -320,6 +320,9 @@ static s32 ixgbe_identify_phy_x550em(struct ixgbe_hw *hw)
 	case IXGBE_DEV_ID_X550EM_X_KX4:
 		hw->phy.type = ixgbe_phy_x550em_kx4;
 		break;
+	case IXGBE_DEV_ID_X550EM_X_XFI:
+		hw->phy.type = ixgbe_phy_x550em_xfi;
+		break;
 	case IXGBE_DEV_ID_X550EM_X_KR:
 	case IXGBE_DEV_ID_X550EM_A_KR:
 	case IXGBE_DEV_ID_X550EM_A_KR_L:
@@ -2231,6 +2234,10 @@ static s32 ixgbe_get_link_capabilities_X550em(struct ixgbe_hw *hw,
 				 IXGBE_LINK_SPEED_2_5GB_FULL |
 				 IXGBE_LINK_SPEED_10GB_FULL;
 			break;
+		case ixgbe_phy_x550em_xfi:
+			*speed = IXGBE_LINK_SPEED_1GB_FULL |
+				 IXGBE_LINK_SPEED_10GB_FULL;
+			break;
 		case ixgbe_phy_sgmii:
 			*speed = IXGBE_LINK_SPEED_1GB_FULL;
 			break;
@@ -3127,6 +3134,12 @@ static s32 ixgbe_init_phy_ops_X550em(struct ixgbe_hw *hw)
 		phy->ops.read_reg = ixgbe_read_phy_reg_x550em;
 		phy->ops.write_reg = ixgbe_write_phy_reg_x550em;
 		break;
+	case ixgbe_phy_x550em_xfi:
+		/* link is managed by HW */
+		phy->ops.setup_link = NULL;
+		phy->ops.read_reg = ixgbe_read_phy_reg_x550em;
+		phy->ops.write_reg = ixgbe_write_phy_reg_x550em;
+		break;
 	case ixgbe_phy_x550em_ext_t:
 		/* Save NW management interface connected on board. This is used
 		 * to determine internal PHY mode
@@ -3180,6 +3193,7 @@ static enum ixgbe_media_type ixgbe_get_media_type_X550em(struct ixgbe_hw *hw)
 		/* Fallthrough */
 	case IXGBE_DEV_ID_X550EM_X_KR:
 	case IXGBE_DEV_ID_X550EM_X_KX4:
+	case IXGBE_DEV_ID_X550EM_X_XFI:
 	case IXGBE_DEV_ID_X550EM_A_KR:
 	case IXGBE_DEV_ID_X550EM_A_KR_L:
 		media_type = ixgbe_media_type_backplane;
-- 
2.12.2

[net-next 06/15] ixgbe: list X553 backplane speeds correctly

[Jeff Kirsher]

From: Don Skidmore <donald.c.skidmore@intel.com>

We forgot to indicate some of the supported speed on the X553
backplane.  This patch attempts to correct for that.

Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Tested-by: Krishneil Singh <krishneil.k.singh@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_type.h |  5 +++++
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 14 ++++++++++++++
 2 files changed, 19 insertions(+)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
index 22869aaec3d4..f5fdd9b17de2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
@@ -3775,6 +3775,11 @@ struct ixgbe_info {
 
 #define IXGBE_NW_MNG_IF_SEL		0x00011178
 #define IXGBE_NW_MNG_IF_SEL_MDIO_ACT		BIT(1)
+#define IXGBE_NW_MNG_IF_SEL_PHY_SPEED_10M	BIT(17)
+#define IXGBE_NW_MNG_IF_SEL_PHY_SPEED_100M	BIT(18)
+#define IXGBE_NW_MNG_IF_SEL_PHY_SPEED_1G	BIT(19)
+#define IXGBE_NW_MNG_IF_SEL_PHY_SPEED_2_5G	BIT(20)
+#define IXGBE_NW_MNG_IF_SEL_PHY_SPEED_10G	BIT(21)
 #define IXGBE_NW_MNG_IF_SEL_ENABLE_10_100M	BIT(23)
 #define IXGBE_NW_MNG_IF_SEL_INT_PHY_MODE	BIT(24)
 #define IXGBE_NW_MNG_IF_SEL_MDIO_PHY_ADD_SHIFT	3
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
index 5127e5e759f2..af579c21e57f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
@@ -2241,6 +2241,20 @@ static s32 ixgbe_get_link_capabilities_X550em(struct ixgbe_hw *hw,
 		case ixgbe_phy_sgmii:
 			*speed = IXGBE_LINK_SPEED_1GB_FULL;
 			break;
+		case ixgbe_phy_x550em_kr:
+			if (hw->mac.type == ixgbe_mac_x550em_a) {
+				/* check different backplane modes */
+				if (hw->phy.nw_mng_if_sel &
+				    IXGBE_NW_MNG_IF_SEL_PHY_SPEED_2_5G) {
+					*speed = IXGBE_LINK_SPEED_2_5GB_FULL;
+					break;
+				} else if (hw->device_id ==
+					   IXGBE_DEV_ID_X550EM_A_KR_L) {
+					*speed = IXGBE_LINK_SPEED_1GB_FULL;
+					break;
+				}
+			}
+			/* fall through */
 		default:
 			*speed = IXGBE_LINK_SPEED_10GB_FULL |
 				 IXGBE_LINK_SPEED_1GB_FULL;
-- 
2.12.2

[net-next 07/15] ixgbe: add default setup_link for x550em_a MAC type

[Jeff Kirsher]

From: Emil Tantilov <emil.s.tantilov@intel.com>

Add default setting for mac->ops.setup_link on x550em_a MAC types.
This fixes a link issue on KR parts.

Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
index af579c21e57f..04d6c744b936 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
@@ -3795,7 +3795,7 @@ static struct ixgbe_mac_operations mac_ops_x550em_a = {
 	.get_media_type		= ixgbe_get_media_type_X550em,
 	.get_san_mac_addr	= NULL,
 	.get_wwn_prefix		= NULL,
-	.setup_link		= NULL, /* defined later */
+	.setup_link		= &ixgbe_setup_mac_link_X540,
 	.get_link_capabilities	= ixgbe_get_link_capabilities_X550em,
 	.get_bus_info		= ixgbe_get_bus_info_X550em,
 	.setup_sfp		= ixgbe_setup_sfp_modules_X550em,
-- 
2.12.2

[net-next 08/15] ixgbe: move num_vfs_macvlans allocation into separate function

[Jeff Kirsher]

From: Emil Tantilov <emil.s.tantilov@intel.com>

Move the code allocating memory for list of MAC addresses that
the VFs can use for MACVLAN into its own function.

Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c | 48 ++++++++++++++++----------
 1 file changed, 29 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
index 39e109da9bd9..d10b25fa0be4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
@@ -46,42 +46,50 @@
 #include "ixgbe_sriov.h"
 
 #ifdef CONFIG_PCI_IOV
-static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
+static inline void ixgbe_alloc_vf_macvlans(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	int num_vf_macvlans, i;
 	struct vf_macvlans *mv_list;
-
-	adapter->flags |= IXGBE_FLAG_SRIOV_ENABLED;
-	e_info(probe, "SR-IOV enabled with %d VFs\n", adapter->num_vfs);
-
-	/* Enable VMDq flag so device will be set in VM mode */
-	adapter->flags |= IXGBE_FLAG_VMDQ_ENABLED;
-	if (!adapter->ring_feature[RING_F_VMDQ].limit)
-		adapter->ring_feature[RING_F_VMDQ].limit = 1;
-	adapter->ring_feature[RING_F_VMDQ].offset = adapter->num_vfs;
+	int num_vf_macvlans, i;
 
 	num_vf_macvlans = hw->mac.num_rar_entries -
-	(IXGBE_MAX_PF_MACVLANS + 1 + adapter->num_vfs);
+			  (IXGBE_MAX_PF_MACVLANS + 1 + adapter->num_vfs);
+	if (!num_vf_macvlans)
+		return;
 
-	adapter->mv_list = mv_list = kcalloc(num_vf_macvlans,
-					     sizeof(struct vf_macvlans),
-					     GFP_KERNEL);
+	mv_list = kcalloc(num_vf_macvlans, sizeof(struct vf_macvlans),
+			  GFP_KERNEL);
 	if (mv_list) {
 		/* Initialize list of VF macvlans */
 		INIT_LIST_HEAD(&adapter->vf_mvs.l);
 		for (i = 0; i < num_vf_macvlans; i++) {
-			mv_list->vf = -1;
-			mv_list->free = true;
-			list_add(&mv_list->l, &adapter->vf_mvs.l);
-			mv_list++;
+			mv_list[i].vf = -1;
+			mv_list[i].free = true;
+			list_add(&mv_list[i].l, &adapter->vf_mvs.l);
 		}
+		adapter->mv_list = mv_list;
 	}
+}
+
+static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+
+	adapter->flags |= IXGBE_FLAG_SRIOV_ENABLED;
+	e_info(probe, "SR-IOV enabled with %d VFs\n", adapter->num_vfs);
+
+	/* Enable VMDq flag so device will be set in VM mode */
+	adapter->flags |= IXGBE_FLAG_VMDQ_ENABLED;
+	if (!adapter->ring_feature[RING_F_VMDQ].limit)
+		adapter->ring_feature[RING_F_VMDQ].limit = 1;
+	adapter->ring_feature[RING_F_VMDQ].offset = adapter->num_vfs;
 
 	/* Initialize default switching mode VEB */
 	IXGBE_WRITE_REG(hw, IXGBE_PFDTXGSWC, IXGBE_PFDTXGSWC_VT_LBEN);
 	adapter->bridge_mode = BRIDGE_MODE_VEB;
 
+	ixgbe_alloc_vf_macvlans(adapter);
+
 	/* If call to enable VFs succeeded then allocate memory
 	 * for per VF control structures.
 	 */
@@ -89,6 +97,8 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 		kcalloc(adapter->num_vfs,
 			sizeof(struct vf_data_storage), GFP_KERNEL);
 	if (adapter->vfinfo) {
+		int i;
+
 		/* limit trafffic classes based on VFs enabled */
 		if ((adapter->hw.mac.type == ixgbe_mac_82599EB) &&
 		    (adapter->num_vfs < 16)) {
-- 
2.12.2

[net-next 09/15] ixgbe: return early instead of wrap block in if statement

[Jeff Kirsher]

From: Emil Tantilov <emil.s.tantilov@intel.com>

Since we exit at the end of the block, we can save a level of
indentation by performing an early return, and make the next several
sections of code more legible, with fewer 80 character line breaks.

Also moved allocating vfinfo at the beginning and the notification
for enabling SRIOV at the end of the function when we know that it
will succeed.

Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c | 85 +++++++++++++-------------
 1 file changed, 41 insertions(+), 44 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
index d10b25fa0be4..16952d33730e 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
@@ -74,9 +74,9 @@ static inline void ixgbe_alloc_vf_macvlans(struct ixgbe_adapter *adapter)
 static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
+	int i;
 
 	adapter->flags |= IXGBE_FLAG_SRIOV_ENABLED;
-	e_info(probe, "SR-IOV enabled with %d VFs\n", adapter->num_vfs);
 
 	/* Enable VMDq flag so device will be set in VM mode */
 	adapter->flags |= IXGBE_FLAG_VMDQ_ENABLED;
@@ -84,60 +84,57 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 		adapter->ring_feature[RING_F_VMDQ].limit = 1;
 	adapter->ring_feature[RING_F_VMDQ].offset = adapter->num_vfs;
 
-	/* Initialize default switching mode VEB */
-	IXGBE_WRITE_REG(hw, IXGBE_PFDTXGSWC, IXGBE_PFDTXGSWC_VT_LBEN);
-	adapter->bridge_mode = BRIDGE_MODE_VEB;
-
-	ixgbe_alloc_vf_macvlans(adapter);
-
 	/* If call to enable VFs succeeded then allocate memory
 	 * for per VF control structures.
 	 */
-	adapter->vfinfo =
-		kcalloc(adapter->num_vfs,
-			sizeof(struct vf_data_storage), GFP_KERNEL);
-	if (adapter->vfinfo) {
-		int i;
-
-		/* limit trafffic classes based on VFs enabled */
-		if ((adapter->hw.mac.type == ixgbe_mac_82599EB) &&
-		    (adapter->num_vfs < 16)) {
-			adapter->dcb_cfg.num_tcs.pg_tcs = MAX_TRAFFIC_CLASS;
-			adapter->dcb_cfg.num_tcs.pfc_tcs = MAX_TRAFFIC_CLASS;
-		} else if (adapter->num_vfs < 32) {
-			adapter->dcb_cfg.num_tcs.pg_tcs = 4;
-			adapter->dcb_cfg.num_tcs.pfc_tcs = 4;
-		} else {
-			adapter->dcb_cfg.num_tcs.pg_tcs = 1;
-			adapter->dcb_cfg.num_tcs.pfc_tcs = 1;
-		}
+	adapter->vfinfo = kcalloc(adapter->num_vfs,
+				  sizeof(struct vf_data_storage), GFP_KERNEL);
+	if (!adapter->vfinfo)
+		return -ENOMEM;
+
+	ixgbe_alloc_vf_macvlans(adapter);
+
+	/* Initialize default switching mode VEB */
+	IXGBE_WRITE_REG(hw, IXGBE_PFDTXGSWC, IXGBE_PFDTXGSWC_VT_LBEN);
+	adapter->bridge_mode = BRIDGE_MODE_VEB;
 
-		/* Disable RSC when in SR-IOV mode */
-		adapter->flags2 &= ~(IXGBE_FLAG2_RSC_CAPABLE |
-				     IXGBE_FLAG2_RSC_ENABLED);
+	/* limit trafffic classes based on VFs enabled */
+	if ((adapter->hw.mac.type == ixgbe_mac_82599EB) &&
+	    (adapter->num_vfs < 16)) {
+		adapter->dcb_cfg.num_tcs.pg_tcs = MAX_TRAFFIC_CLASS;
+		adapter->dcb_cfg.num_tcs.pfc_tcs = MAX_TRAFFIC_CLASS;
+	} else if (adapter->num_vfs < 32) {
+		adapter->dcb_cfg.num_tcs.pg_tcs = 4;
+		adapter->dcb_cfg.num_tcs.pfc_tcs = 4;
+	} else {
+		adapter->dcb_cfg.num_tcs.pg_tcs = 1;
+		adapter->dcb_cfg.num_tcs.pfc_tcs = 1;
+	}
 
-		for (i = 0; i < adapter->num_vfs; i++) {
-			/* enable spoof checking for all VFs */
-			adapter->vfinfo[i].spoofchk_enabled = true;
+	/* Disable RSC when in SR-IOV mode */
+	adapter->flags2 &= ~(IXGBE_FLAG2_RSC_CAPABLE |
+			     IXGBE_FLAG2_RSC_ENABLED);
 
-			/* We support VF RSS querying only for 82599 and x540
-			 * devices at the moment. These devices share RSS
-			 * indirection table and RSS hash key with PF therefore
-			 * we want to disable the querying by default.
-			 */
-			adapter->vfinfo[i].rss_query_enabled = 0;
+	for (i = 0; i < adapter->num_vfs; i++) {
+		/* enable spoof checking for all VFs */
+		adapter->vfinfo[i].spoofchk_enabled = true;
 
-			/* Untrust all VFs */
-			adapter->vfinfo[i].trusted = false;
+		/* We support VF RSS querying only for 82599 and x540
+		 * devices at the moment. These devices share RSS
+		 * indirection table and RSS hash key with PF therefore
+		 * we want to disable the querying by default.
+		 */
+		adapter->vfinfo[i].rss_query_enabled = 0;
 
-			/* set the default xcast mode */
-			adapter->vfinfo[i].xcast_mode = IXGBEVF_XCAST_MODE_NONE;
-		}
+		/* Untrust all VFs */
+		adapter->vfinfo[i].trusted = false;
 
-		return 0;
+		/* set the default xcast mode */
+		adapter->vfinfo[i].xcast_mode = IXGBEVF_XCAST_MODE_NONE;
 	}
 
-	return -ENOMEM;
+	e_info(probe, "SR-IOV enabled with %d VFs\n", adapter->num_vfs);
+	return 0;
 }
 
 /**
-- 
2.12.2

[net-next 10/15] ixgbe: do not use adapter->num_vfs when setting VFs via module parameter

[Jeff Kirsher]

From: Emil Tantilov <emil.s.tantilov@intel.com>

Avoid setting adapter->num_vfs early in the init code path when
using the max_vfs module parameter by passing it to ixgbe_enable_sriov()
as a function parameter.

This fixes an issue where if we failed to allocate vfinfo in
__ixgbe_enable_sriov() the driver will crash with NULL pointer in
ixgbe_disable_sriov() when attempting to free the vfinfo struct based
on adapter->num_vfs. Also it cleans up the assignment of adapter->num_vfs
since now it will only be set in __ixgbe_enable_sriov() and cleared in
ixgbe_disable_sriov().

Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c  |  6 ++--
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c | 50 +++++++++++++-------------
 drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h |  2 +-
 3 files changed, 28 insertions(+), 30 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index e250c1d54a11..dcc8a5c26276 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -5955,10 +5955,8 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter,
 	/* assign number of SR-IOV VFs */
 	if (hw->mac.type != ixgbe_mac_82598EB) {
 		if (max_vfs > IXGBE_MAX_VFS_DRV_LIMIT) {
-			adapter->num_vfs = 0;
+			max_vfs = 0;
 			e_dev_warn("max_vfs parameter out of range. Not assigning any SR-IOV VFs\n");
-		} else {
-			adapter->num_vfs = max_vfs;
 		}
 	}
 #endif /* CONFIG_PCI_IOV */
@@ -9821,7 +9819,7 @@ static int ixgbe_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	ixgbe_init_mbx_params_pf(hw);
 	hw->mbx.ops = ii->mbx_ops;
 	pci_sriov_set_totalvfs(pdev, IXGBE_MAX_VFS_DRV_LIMIT);
-	ixgbe_enable_sriov(adapter);
+	ixgbe_enable_sriov(adapter, max_vfs);
 skip_sriov:
 
 #endif
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
index 16952d33730e..102ca937ddb4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
@@ -46,14 +46,15 @@
 #include "ixgbe_sriov.h"
 
 #ifdef CONFIG_PCI_IOV
-static inline void ixgbe_alloc_vf_macvlans(struct ixgbe_adapter *adapter)
+static inline void ixgbe_alloc_vf_macvlans(struct ixgbe_adapter *adapter,
+					   unsigned int num_vfs)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	struct vf_macvlans *mv_list;
 	int num_vf_macvlans, i;
 
 	num_vf_macvlans = hw->mac.num_rar_entries -
-			  (IXGBE_MAX_PF_MACVLANS + 1 + adapter->num_vfs);
+			  (IXGBE_MAX_PF_MACVLANS + 1 + num_vfs);
 	if (!num_vf_macvlans)
 		return;
 
@@ -71,7 +72,8 @@ static inline void ixgbe_alloc_vf_macvlans(struct ixgbe_adapter *adapter)
 	}
 }
 
-static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
+static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter,
+				unsigned int num_vfs)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	int i;
@@ -82,28 +84,27 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 	adapter->flags |= IXGBE_FLAG_VMDQ_ENABLED;
 	if (!adapter->ring_feature[RING_F_VMDQ].limit)
 		adapter->ring_feature[RING_F_VMDQ].limit = 1;
-	adapter->ring_feature[RING_F_VMDQ].offset = adapter->num_vfs;
 
-	/* If call to enable VFs succeeded then allocate memory
-	 * for per VF control structures.
-	 */
-	adapter->vfinfo = kcalloc(adapter->num_vfs,
-				  sizeof(struct vf_data_storage), GFP_KERNEL);
+	/* Allocate memory for per VF control structures */
+	adapter->vfinfo = kcalloc(num_vfs, sizeof(struct vf_data_storage),
+				  GFP_KERNEL);
 	if (!adapter->vfinfo)
 		return -ENOMEM;
 
-	ixgbe_alloc_vf_macvlans(adapter);
+	adapter->num_vfs = num_vfs;
+
+	ixgbe_alloc_vf_macvlans(adapter, num_vfs);
+	adapter->ring_feature[RING_F_VMDQ].offset = num_vfs;
 
 	/* Initialize default switching mode VEB */
 	IXGBE_WRITE_REG(hw, IXGBE_PFDTXGSWC, IXGBE_PFDTXGSWC_VT_LBEN);
 	adapter->bridge_mode = BRIDGE_MODE_VEB;
 
 	/* limit trafffic classes based on VFs enabled */
-	if ((adapter->hw.mac.type == ixgbe_mac_82599EB) &&
-	    (adapter->num_vfs < 16)) {
+	if ((adapter->hw.mac.type == ixgbe_mac_82599EB) && (num_vfs < 16)) {
 		adapter->dcb_cfg.num_tcs.pg_tcs = MAX_TRAFFIC_CLASS;
 		adapter->dcb_cfg.num_tcs.pfc_tcs = MAX_TRAFFIC_CLASS;
-	} else if (adapter->num_vfs < 32) {
+	} else if (num_vfs < 32) {
 		adapter->dcb_cfg.num_tcs.pg_tcs = 4;
 		adapter->dcb_cfg.num_tcs.pfc_tcs = 4;
 	} else {
@@ -115,7 +116,7 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 	adapter->flags2 &= ~(IXGBE_FLAG2_RSC_CAPABLE |
 			     IXGBE_FLAG2_RSC_ENABLED);
 
-	for (i = 0; i < adapter->num_vfs; i++) {
+	for (i = 0; i < num_vfs; i++) {
 		/* enable spoof checking for all VFs */
 		adapter->vfinfo[i].spoofchk_enabled = true;
 
@@ -133,7 +134,7 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 		adapter->vfinfo[i].xcast_mode = IXGBEVF_XCAST_MODE_NONE;
 	}
 
-	e_info(probe, "SR-IOV enabled with %d VFs\n", adapter->num_vfs);
+	e_info(probe, "SR-IOV enabled with %d VFs\n", num_vfs);
 	return 0;
 }
 
@@ -172,12 +173,13 @@ static void ixgbe_get_vfs(struct ixgbe_adapter *adapter)
 /* Note this function is called when the user wants to enable SR-IOV
  * VFs using the now deprecated module parameter
  */
-void ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
+void ixgbe_enable_sriov(struct ixgbe_adapter *adapter, unsigned int max_vfs)
 {
 	int pre_existing_vfs = 0;
+	unsigned int num_vfs;
 
 	pre_existing_vfs = pci_num_vf(adapter->pdev);
-	if (!pre_existing_vfs && !adapter->num_vfs)
+	if (!pre_existing_vfs && !max_vfs)
 		return;
 
 	/* If there are pre-existing VFs then we have to force
@@ -187,7 +189,7 @@ void ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 	 * have been created via the new PCI SR-IOV sysfs interface.
 	 */
 	if (pre_existing_vfs) {
-		adapter->num_vfs = pre_existing_vfs;
+		num_vfs = pre_existing_vfs;
 		dev_warn(&adapter->pdev->dev,
 			 "Virtual Functions already enabled for this device - Please reload all VF drivers to avoid spoofed packet errors\n");
 	} else {
@@ -199,17 +201,16 @@ void ixgbe_enable_sriov(struct ixgbe_adapter *adapter)
 		 * physical function.  If the user requests greater than
 		 * 63 VFs then it is an error - reset to default of zero.
 		 */
-		adapter->num_vfs = min_t(unsigned int, adapter->num_vfs, IXGBE_MAX_VFS_DRV_LIMIT);
+		num_vfs = min_t(unsigned int, max_vfs, IXGBE_MAX_VFS_DRV_LIMIT);
 
-		err = pci_enable_sriov(adapter->pdev, adapter->num_vfs);
+		err = pci_enable_sriov(adapter->pdev, num_vfs);
 		if (err) {
 			e_err(probe, "Failed to enable PCI sriov: %d\n", err);
-			adapter->num_vfs = 0;
 			return;
 		}
 	}
 
-	if (!__ixgbe_enable_sriov(adapter)) {
+	if (!__ixgbe_enable_sriov(adapter, num_vfs)) {
 		ixgbe_get_vfs(adapter);
 		return;
 	}
@@ -347,13 +348,12 @@ static int ixgbe_pci_sriov_enable(struct pci_dev *dev, int num_vfs)
 			return -EPERM;
 		}
 	}
-	adapter->num_vfs = num_vfs;
 
-	err = __ixgbe_enable_sriov(adapter);
+	err = __ixgbe_enable_sriov(adapter, num_vfs);
 	if (err)
 		return  err;
 
-	for (i = 0; i < adapter->num_vfs; i++)
+	for (i = 0; i < num_vfs; i++)
 		ixgbe_vf_configuration(dev, (i | 0x10000000));
 
 	/* reset before enabling SRIOV to avoid mailbox issues */
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
index 3166fd164e51..cf67b9b18ed7 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h
@@ -59,7 +59,7 @@ int ixgbe_ndo_get_vf_config(struct net_device *netdev,
 void ixgbe_check_vf_rate_limit(struct ixgbe_adapter *adapter);
 int ixgbe_disable_sriov(struct ixgbe_adapter *adapter);
 #ifdef CONFIG_PCI_IOV
-void ixgbe_enable_sriov(struct ixgbe_adapter *adapter);
+void ixgbe_enable_sriov(struct ixgbe_adapter *adapter, unsigned int max_vfs);
 #endif
 int ixgbe_pci_sriov_configure(struct pci_dev *dev, int num_vfs);
 
-- 
2.12.2

[net-next 11/15] ixgbe: Remove unused define

[Jeff Kirsher]

From: Don Skidmore <donald.c.skidmore@intel.com>

Remove the Marvell 1145 PHY define as we have never had a device that
supports it and have no plan to in the future.  The existence of this
define has caused confusing on whether or not this PHY was supported
by ixgbe.

Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_type.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
index f5fdd9b17de2..2f06e4d9208d 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
@@ -1388,9 +1388,6 @@ struct ixgbe_thermal_sensor_data {
 #define ATH_PHY_ID       0x03429050
 #define AQ_FW_REV        0x20
 
-/* PHY Types */
-#define IXGBE_M88E1145_E_PHY_ID  0x01410CD0
-
 /* Special PHY Init Routine */
 #define IXGBE_PHY_INIT_OFFSET_NL 0x002B
 #define IXGBE_PHY_INIT_END_NL    0xFFFF
-- 
2.12.2

[net-next 12/15] ixgbevf: use new api ethtool_{get|set}_link_ksettings

[Jeff Kirsher]

From: Philippe Reynes <tremyfr@gmail.com>

The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.

As I don't have the hardware, I'd be very pleased if
someone may test this patch.

Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbevf/ethtool.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbevf/ethtool.c b/drivers/net/ethernet/intel/ixgbevf/ethtool.c
index 1f6c0ecd50bb..6bf740945260 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ethtool.c
@@ -91,18 +91,18 @@ static const char ixgbe_gstrings_test[][ETH_GSTRING_LEN] = {
 
 #define IXGBEVF_TEST_LEN (sizeof(ixgbe_gstrings_test) / ETH_GSTRING_LEN)
 
-static int ixgbevf_get_settings(struct net_device *netdev,
-				struct ethtool_cmd *ecmd)
+static int ixgbevf_get_link_ksettings(struct net_device *netdev,
+				      struct ethtool_link_ksettings *cmd)
 {
 	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
 	u32 link_speed = 0;
 	bool link_up;
 
-	ecmd->supported = SUPPORTED_10000baseT_Full;
-	ecmd->autoneg = AUTONEG_DISABLE;
-	ecmd->transceiver = XCVR_DUMMY1;
-	ecmd->port = -1;
+	ethtool_link_ksettings_zero_link_mode(cmd, supported);
+	ethtool_link_ksettings_add_link_mode(cmd, supported, 10000baseT_Full);
+	cmd->base.autoneg = AUTONEG_DISABLE;
+	cmd->base.port = -1;
 
 	hw->mac.get_link_status = 1;
 	hw->mac.ops.check_link(hw, &link_speed, &link_up, false);
@@ -122,11 +122,11 @@ static int ixgbevf_get_settings(struct net_device *netdev,
 			break;
 		}
 
-		ethtool_cmd_speed_set(ecmd, speed);
-		ecmd->duplex = DUPLEX_FULL;
+		cmd->base.speed = speed;
+		cmd->base.duplex = DUPLEX_FULL;
 	} else {
-		ethtool_cmd_speed_set(ecmd, SPEED_UNKNOWN);
-		ecmd->duplex = DUPLEX_UNKNOWN;
+		cmd->base.speed = SPEED_UNKNOWN;
+		cmd->base.duplex = DUPLEX_UNKNOWN;
 	}
 
 	return 0;
@@ -885,7 +885,6 @@ static int ixgbevf_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 }
 
 static const struct ethtool_ops ixgbevf_ethtool_ops = {
-	.get_settings		= ixgbevf_get_settings,
 	.get_drvinfo		= ixgbevf_get_drvinfo,
 	.get_regs_len		= ixgbevf_get_regs_len,
 	.get_regs		= ixgbevf_get_regs,
@@ -905,6 +904,7 @@ static const struct ethtool_ops ixgbevf_ethtool_ops = {
 	.get_rxfh_indir_size	= ixgbevf_get_rxfh_indir_size,
 	.get_rxfh_key_size	= ixgbevf_get_rxfh_key_size,
 	.get_rxfh		= ixgbevf_get_rxfh,
+	.get_link_ksettings	= ixgbevf_get_link_ksettings,
 };
 
 void ixgbevf_set_ethtool_ops(struct net_device *netdev)
-- 
2.12.2

[net-next 13/15] ixgbe: add check for VETO bit when configuring link for KR

[Jeff Kirsher]

From: Tony Nguyen <anthony.l.nguyen@intel.com>

We did not have a check in place for MMNGC.MNG_VETO when setting up link
on X550EM_X KR devices which resulted in link loss for the BMC when
loading the driver.

This patch adds a check for ixgbe_check_reset_blocked() in setup_link()
since in that case there is no PHY reset function.

Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
index 04d6c744b936..2658394599e4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_x550.c
@@ -2526,6 +2526,9 @@ static s32 ixgbe_setup_kr_x550em(struct ixgbe_hw *hw)
 	if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_2_5GB_FULL)
 		return 0;
 
+	if (ixgbe_check_reset_blocked(hw))
+		return 0;
+
 	return ixgbe_setup_kr_speed_x550em(hw, hw->phy.autoneg_advertised);
 }
 
-- 
2.12.2

[net-next 14/15] ixgbe: Add support for maximum headroom when using build_skb

[Jeff Kirsher]

From: Alexander Duyck <alexander.h.duyck@intel.com>

This patch increases the headroom allocated when using build_skb on a
system with 4K pages.  Specifically the breakdown of headroom versus cache
size is as follows:
    L1 Cache Size           Headroom
    64                      192
    64, NET_IP_ALIGN == 2   194
    128                     128
    128, NET_IP_ALIGN == 2  130
    256                     512
    256, NET_IP_ALIGN == 2  258

I stopped at supporting only a cache line size of 256 as that was the
largest cache size I could find supported in the kernel.

With this we are guaranteeing at least 128 bytes of headroom to spare in
the frame.  This should be enough for us to insert a couple of IPv6 headers
if needed which is likely enough room for anything XDP should need.

I'm leaving the padding for systems with pages larger than 4K unmodified
for now.  XDP currently isn't really setup to work on those types of
systems so we can cross that bridge when we get there.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe.h      | 55 ++++++++++++++++++++++++---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |  6 +--
 2 files changed, 53 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index b1ecc2627a5a..656ca8f69768 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -86,17 +86,62 @@
 
 /* Supported Rx Buffer Sizes */
 #define IXGBE_RXBUFFER_256    256  /* Used for skb receive header */
+#define IXGBE_RXBUFFER_1536  1536
 #define IXGBE_RXBUFFER_2K    2048
 #define IXGBE_RXBUFFER_3K    3072
 #define IXGBE_RXBUFFER_4K    4096
 #define IXGBE_MAX_RXBUFFER  16384  /* largest size for a single descriptor */
 
-#define IXGBE_SKB_PAD		(NET_SKB_PAD + NET_IP_ALIGN)
+/* Attempt to maximize the headroom available for incoming frames.  We
+ * use a 2K buffer for receives and need 1536/1534 to store the data for
+ * the frame.  This leaves us with 512 bytes of room.  From that we need
+ * to deduct the space needed for the shared info and the padding needed
+ * to IP align the frame.
+ *
+ * Note: For cache line sizes 256 or larger this value is going to end
+ *	 up negative.  In these cases we should fall back to the 3K
+ *	 buffers.
+ */
 #if (PAGE_SIZE < 8192)
-#define IXGBE_MAX_FRAME_BUILD_SKB \
-	(SKB_WITH_OVERHEAD(IXGBE_RXBUFFER_2K) - IXGBE_SKB_PAD)
+#define IXGBE_MAX_2K_FRAME_BUILD_SKB (IXGBE_RXBUFFER_1536 - NET_IP_ALIGN)
+#define IXGBE_2K_TOO_SMALL_WITH_PADDING \
+((NET_SKB_PAD + IXGBE_RXBUFFER_1536) > SKB_WITH_OVERHEAD(IXGBE_RXBUFFER_2K))
+
+static inline int ixgbe_compute_pad(int rx_buf_len)
+{
+	int page_size, pad_size;
+
+	page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2);
+	pad_size = SKB_WITH_OVERHEAD(page_size) - rx_buf_len;
+
+	return pad_size;
+}
+
+static inline int ixgbe_skb_pad(void)
+{
+	int rx_buf_len;
+
+	/* If a 2K buffer cannot handle a standard Ethernet frame then
+	 * optimize padding for a 3K buffer instead of a 1.5K buffer.
+	 *
+	 * For a 3K buffer we need to add enough padding to allow for
+	 * tailroom due to NET_IP_ALIGN possibly shifting us out of
+	 * cache-line alignment.
+	 */
+	if (IXGBE_2K_TOO_SMALL_WITH_PADDING)
+		rx_buf_len = IXGBE_RXBUFFER_3K + SKB_DATA_ALIGN(NET_IP_ALIGN);
+	else
+		rx_buf_len = IXGBE_RXBUFFER_1536;
+
+	/* if needed make room for NET_IP_ALIGN */
+	rx_buf_len -= NET_IP_ALIGN;
+
+	return ixgbe_compute_pad(rx_buf_len);
+}
+
+#define IXGBE_SKB_PAD	ixgbe_skb_pad()
 #else
-#define IXGBE_MAX_FRAME_BUILD_SKB IXGBE_RXBUFFER_2K
+#define IXGBE_SKB_PAD	(NET_SKB_PAD + NET_IP_ALIGN)
 #endif
 
 /*
@@ -361,7 +406,7 @@ static inline unsigned int ixgbe_rx_bufsz(struct ixgbe_ring *ring)
 		return IXGBE_RXBUFFER_3K;
 #if (PAGE_SIZE < 8192)
 	if (ring_uses_build_skb(ring))
-		return IXGBE_MAX_FRAME_BUILD_SKB;
+		return IXGBE_MAX_2K_FRAME_BUILD_SKB;
 #endif
 	return IXGBE_RXBUFFER_2K;
 }
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index dcc8a5c26276..536dd9b1ad97 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -3813,7 +3813,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
 		/* Limit the maximum frame size so we don't overrun the skb */
 		if (ring_uses_build_skb(ring) &&
 		    !test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
-			rxdctl |= IXGBE_MAX_FRAME_BUILD_SKB |
+			rxdctl |= IXGBE_MAX_2K_FRAME_BUILD_SKB |
 				  IXGBE_RXDCTL_RLPML_EN;
 #endif
 	}
@@ -3983,8 +3983,8 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
 		if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
 			set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
 
-		if ((max_frame > (ETH_FRAME_LEN + ETH_FCS_LEN)) ||
-		    (max_frame > IXGBE_MAX_FRAME_BUILD_SKB))
+		if (IXGBE_2K_TOO_SMALL_WITH_PADDING ||
+		    (max_frame > (ETH_FRAME_LEN + ETH_FCS_LEN)))
 			set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
 #endif
 	}
-- 
2.12.2

[net-next 15/15] ixgbe: Fix output from ixgbe_dump

[Jeff Kirsher]

From: Alexander Duyck <alexander.h.duyck@intel.com>

I just found that when we had changed the Rx path to check for length
instead of the DD bit we introduced an issue in ixgbe_dump since we were no
longer clearing the status bits.

To correct this I am updating ixgbe_dump to look for the length bits in the
descriptor since that is what we are using in the Rx path.

Fixes: c3630cc40b4f ("ixgbe: Use length to determine if descriptor is done")
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 536dd9b1ad97..afff2ca7f8c0 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -606,7 +606,6 @@ static void ixgbe_dump(struct ixgbe_adapter *adapter)
 	struct ixgbe_ring *rx_ring;
 	union ixgbe_adv_rx_desc *rx_desc;
 	struct ixgbe_rx_buffer *rx_buffer_info;
-	u32 staterr;
 	int i = 0;
 
 	if (!netif_msg_hw(adapter))
@@ -827,8 +826,7 @@ static void ixgbe_dump(struct ixgbe_adapter *adapter)
 			rx_buffer_info = &rx_ring->rx_buffer_info[i];
 			rx_desc = IXGBE_RX_DESC(rx_ring, i);
 			u0 = (struct my_u0 *)rx_desc;
-			staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
-			if (staterr & IXGBE_RXD_STAT_DD) {
+			if (rx_desc->wb.upper.length) {
 				/* Descriptor Done */
 				pr_info("RWB[0x%03X]     %016llX %016llX ---------------- %p%s\n",
 					i,
-- 
2.12.2

Re: [net-next 00/15][pull request] 10GbE Intel Wired LAN Driver Updates 2017-04-18

[Alexei Starovoitov]

On Tue, Apr 18, 2017 at 04:01:50PM -0700, Jeff Kirsher wrote:
> The following are changes since commit 4116c97689b9b1732ac5b68afd922406f9fc842e:
>   Merge branch 'ftgmac100-batch5-features'
> and are available in the git repository at:
>   git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue 10GbE

Nothing against this set of commits,
but I saw ixgbe xdp support in there pending for few months.
Now it's gone. What happened?

Re: [net-next 00/15][pull request] 10GbE Intel Wired LAN Driver Updates 2017-04-18

[Jeff Kirsher]

[-- Attachment #1: Type: text/plain, Size: 766 bytes --]

On Wed, 2017-04-19 at 22:25 -0700, Alexei Starovoitov wrote:
> On Tue, Apr 18, 2017 at 04:01:50PM -0700, Jeff Kirsher wrote:
> > The following are changes since commit
> > 4116c97689b9b1732ac5b68afd922406f9fc842e:
> >    Merge branch 'ftgmac100-batch5-features'
> > and are available in the git repository at:
> >    git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-
> > queue 10GbE
> 
> Nothing against this set of commits,
> but I saw ixgbe xdp support in there pending for few months.
> Now it's gone. What happened?

XDP is in the next series of patches, per Dave, we need to keep patch
series smaller to help in revue.  Next series will contain what you are
looking for, just need to get this up before I can push the next
series.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [net-next 00/15][pull request] 10GbE Intel Wired LAN Driver Updates 2017-04-18

[David Miller]

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Tue, 18 Apr 2017 16:01:50 -0700

> This series contains updates to mainly ixgbe with only one ixgbevf
> change.

Pulled, thanks Jeff.