[net-next 0/6][pull request] Intel Wired LAN Driver Updates

[net-next 0/6][pull request] Intel Wired LAN Driver Updates

[Jeff Kirsher]

This series contains updates to ixgbe & ixgbevf.

The following are changes since commit 5abf7f7e0f6bdbfcac737f636497d7016d9507eb:
  ipv4: fix rcu splat
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master

Alexander Duyck (6):
  ixgbe: Ping the VFs on link status change to trigger link change
  ixgbe: Handle failures in the ixgbe_setup_rx/tx_resources calls
  ixgbe: Move configuration of set_real_num_rx/tx_queues into open
  ixgbe: Update the logic for ixgbe_cache_ring_dcb and DCB RSS
    configuration
  ixgbe: Cleanup logic for MRQC and MTQC configuration
  ixgbevf: Update descriptor macros to accept pointers and drop _ADV
    suffix

 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c      |  138 ++++++---------
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c     |  190 +++++++++++++--------
 drivers/net/ethernet/intel/ixgbevf/ixgbevf.h      |   12 +-
 drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c |   18 +-
 4 files changed, 183 insertions(+), 175 deletions(-)

-- 
1.7.10.4

[net-next 1/6] ixgbe: Ping the VFs on link status change to trigger link change

[Jeff Kirsher]

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

When the link status changes on the PF we need to notify the VFs. In order
to do this we should ping all of the VFs in order to trigger a link status
change on them as well.

This fixes issues in which the PF would reset, but the VF didn't because the
NAK flag was not set in the VF mailbox.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |    6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index ee230f5..17f46f0 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -5390,6 +5390,9 @@ static void ixgbe_watchdog_link_is_up(struct ixgbe_adapter *adapter)
 
 	netif_carrier_on(netdev);
 	ixgbe_check_vf_rate_limit(adapter);
+
+	/* ping all the active vfs to let them know link has changed */
+	ixgbe_ping_all_vfs(adapter);
 }
 
 /**
@@ -5419,6 +5422,9 @@ static void ixgbe_watchdog_link_is_down(struct ixgbe_adapter *adapter)
 
 	e_info(drv, "NIC Link is Down\n");
 	netif_carrier_off(netdev);
+
+	/* ping all the active vfs to let them know link has changed */
+	ixgbe_ping_all_vfs(adapter);
 }
 
 /**
-- 
1.7.10.4

[net-next 2/6] ixgbe: Handle failures in the ixgbe_setup_rx/tx_resources calls

[Jeff Kirsher]

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

Previously we were exiting without cleaning up the memory internally on the
ixgbe_setup_rx_resources and ixgbe_setup_tx_resources calls.  Instead of
forcing the caller to clean things up for us we should instead just unwind
the rings and free the memory as we go.  This way we can more gracefully
clean up the rings in the event of an allocation failure.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |   20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 17f46f0..373342f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -4549,10 +4549,16 @@ static int ixgbe_setup_all_tx_resources(struct ixgbe_adapter *adapter)
 		err = ixgbe_setup_tx_resources(adapter->tx_ring[i]);
 		if (!err)
 			continue;
+
 		e_err(probe, "Allocation for Tx Queue %u failed\n", i);
-		break;
+		goto err_setup_tx;
 	}
 
+	return 0;
+err_setup_tx:
+	/* rewind the index freeing the rings as we go */
+	while (i--)
+		ixgbe_free_tx_resources(adapter->tx_ring[i]);
 	return err;
 }
 
@@ -4627,10 +4633,16 @@ static int ixgbe_setup_all_rx_resources(struct ixgbe_adapter *adapter)
 		err = ixgbe_setup_rx_resources(adapter->rx_ring[i]);
 		if (!err)
 			continue;
+
 		e_err(probe, "Allocation for Rx Queue %u failed\n", i);
-		break;
+		goto err_setup_rx;
 	}
 
+	return 0;
+err_setup_rx:
+	/* rewind the index freeing the rings as we go */
+	while (i--)
+		ixgbe_free_rx_resources(adapter->rx_ring[i]);
 	return err;
 }
 
@@ -4791,10 +4803,10 @@ static int ixgbe_open(struct net_device *netdev)
 	return 0;
 
 err_req_irq:
-err_setup_rx:
 	ixgbe_free_all_rx_resources(adapter);
-err_setup_tx:
+err_setup_rx:
 	ixgbe_free_all_tx_resources(adapter);
+err_setup_tx:
 	ixgbe_reset(adapter);
 
 	return err;
-- 
1.7.10.4

[net-next 3/6] ixgbe: Move configuration of set_real_num_rx/tx_queues into open

[Jeff Kirsher]

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

It makes much more sense for us to configure the real number of Tx and Rx
queues in the ixgbe_open call than it does in ixgbe_set_num_queues.  By
setting the number in ixgbe_open we can avoid a number of unecessary
updates and only have to make the calls once.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c  |   58 ++++++-------------------
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |   36 ++++++++++-----
 2 files changed, 38 insertions(+), 56 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
index d308e71..c03d771 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
@@ -349,7 +349,7 @@ static bool ixgbe_set_rss_queues(struct ixgbe_adapter *adapter)
  * fallthrough conditions.
  *
  **/
-static int ixgbe_set_num_queues(struct ixgbe_adapter *adapter)
+static void ixgbe_set_num_queues(struct ixgbe_adapter *adapter)
 {
 	/* Start with base case */
 	adapter->num_rx_queues = 1;
@@ -358,29 +358,14 @@ static int ixgbe_set_num_queues(struct ixgbe_adapter *adapter)
 	adapter->num_rx_queues_per_pool = 1;
 
 	if (ixgbe_set_sriov_queues(adapter))
-		goto done;
+		return;
 
 #ifdef CONFIG_IXGBE_DCB
 	if (ixgbe_set_dcb_queues(adapter))
-		goto done;
+		return;
 
 #endif
-	if (ixgbe_set_rss_queues(adapter))
-		goto done;
-
-	/* fallback to base case */
-	adapter->num_rx_queues = 1;
-	adapter->num_tx_queues = 1;
-
-done:
-	if ((adapter->netdev->reg_state == NETREG_UNREGISTERED) ||
-	    (adapter->netdev->reg_state == NETREG_UNREGISTERING))
-		return 0;
-
-	/* Notify the stack of the (possibly) reduced queue counts. */
-	netif_set_real_num_tx_queues(adapter->netdev, adapter->num_tx_queues);
-	return netif_set_real_num_rx_queues(adapter->netdev,
-					    adapter->num_rx_queues);
+	ixgbe_set_rss_queues(adapter);
 }
 
 static void ixgbe_acquire_msix_vectors(struct ixgbe_adapter *adapter,
@@ -710,11 +695,10 @@ static void ixgbe_reset_interrupt_capability(struct ixgbe_adapter *adapter)
  * Attempt to configure the interrupts using the best available
  * capabilities of the hardware and the kernel.
  **/
-static int ixgbe_set_interrupt_capability(struct ixgbe_adapter *adapter)
+static void ixgbe_set_interrupt_capability(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	int err = 0;
-	int vector, v_budget;
+	int vector, v_budget, err;
 
 	/*
 	 * It's easy to be greedy for MSI-X vectors, but it really
@@ -747,7 +731,7 @@ static int ixgbe_set_interrupt_capability(struct ixgbe_adapter *adapter)
 		ixgbe_acquire_msix_vectors(adapter, v_budget);
 
 		if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED)
-			goto out;
+			return;
 	}
 
 	adapter->flags &= ~IXGBE_FLAG_DCB_ENABLED;
@@ -762,25 +746,17 @@ static int ixgbe_set_interrupt_capability(struct ixgbe_adapter *adapter)
 	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
 		ixgbe_disable_sriov(adapter);
 
-	err = ixgbe_set_num_queues(adapter);
-	if (err)
-		return err;
-
+	ixgbe_set_num_queues(adapter);
 	adapter->num_q_vectors = 1;
 
 	err = pci_enable_msi(adapter->pdev);
-	if (!err) {
-		adapter->flags |= IXGBE_FLAG_MSI_ENABLED;
-	} else {
+	if (err) {
 		netif_printk(adapter, hw, KERN_DEBUG, adapter->netdev,
 			     "Unable to allocate MSI interrupt, "
 			     "falling back to legacy.  Error: %d\n", err);
-		/* reset err */
-		err = 0;
+		return;
 	}
-
-out:
-	return err;
+	adapter->flags |= IXGBE_FLAG_MSI_ENABLED;
 }
 
 /**
@@ -798,15 +774,10 @@ int ixgbe_init_interrupt_scheme(struct ixgbe_adapter *adapter)
 	int err;
 
 	/* Number of supported queues */
-	err = ixgbe_set_num_queues(adapter);
-	if (err)
-		return err;
+	ixgbe_set_num_queues(adapter);
 
-	err = ixgbe_set_interrupt_capability(adapter);
-	if (err) {
-		e_dev_err("Unable to setup interrupt capabilities\n");
-		goto err_set_interrupt;
-	}
+	/* Set interrupt mode */
+	ixgbe_set_interrupt_capability(adapter);
 
 	err = ixgbe_alloc_q_vectors(adapter);
 	if (err) {
@@ -826,7 +797,6 @@ int ixgbe_init_interrupt_scheme(struct ixgbe_adapter *adapter)
 
 err_alloc_q_vectors:
 	ixgbe_reset_interrupt_capability(adapter);
-err_set_interrupt:
 	return err;
 }
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 373342f..7f2aa22 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -4798,10 +4798,26 @@ static int ixgbe_open(struct net_device *netdev)
 	if (err)
 		goto err_req_irq;
 
+	/* Notify the stack of the actual queue counts. */
+	err = netif_set_real_num_tx_queues(netdev,
+					   adapter->num_rx_pools > 1 ? 1 :
+					   adapter->num_tx_queues);
+	if (err)
+		goto err_set_queues;
+
+
+	err = netif_set_real_num_rx_queues(netdev,
+					   adapter->num_rx_pools > 1 ? 1 :
+					   adapter->num_rx_queues);
+	if (err)
+		goto err_set_queues;
+
 	ixgbe_up_complete(adapter);
 
 	return 0;
 
+err_set_queues:
+	ixgbe_free_irq(adapter);
 err_req_irq:
 	ixgbe_free_all_rx_resources(adapter);
 err_setup_rx:
@@ -4864,23 +4880,19 @@ static int ixgbe_resume(struct pci_dev *pdev)
 
 	pci_wake_from_d3(pdev, false);
 
-	rtnl_lock();
-	err = ixgbe_init_interrupt_scheme(adapter);
-	rtnl_unlock();
-	if (err) {
-		e_dev_err("Cannot initialize interrupts for device\n");
-		return err;
-	}
-
 	ixgbe_reset(adapter);
 
 	IXGBE_WRITE_REG(&adapter->hw, IXGBE_WUS, ~0);
 
-	if (netif_running(netdev)) {
+	rtnl_lock();
+	err = ixgbe_init_interrupt_scheme(adapter);
+	if (!err && netif_running(netdev))
 		err = ixgbe_open(netdev);
-		if (err)
-			return err;
-	}
+
+	rtnl_unlock();
+
+	if (err)
+		return err;
 
 	netif_device_attach(netdev);
 
-- 
1.7.10.4

[net-next 4/6] ixgbe: Update the logic for ixgbe_cache_ring_dcb and DCB RSS configuration

[Jeff Kirsher]

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

This change cleans up some of the logic in an attempt to try and simplify
things for how we are configuring DCB w/ RSS.

In this patch I basically did 3 things.  I updated the logic for getting
the first register index.  I applied the fact that all TCs get the same
number of queues to simplify the looping logic in caching the DCB ring
register.  Finally I updated how we configure the RQTC register to match
the fact that all TCs are assigned the same number of queues.

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c  |   80 ++++++++++++-------------
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |   12 ++--
 2 files changed, 42 insertions(+), 50 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
index c03d771..4c3822f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
@@ -42,42 +42,37 @@ static void ixgbe_get_first_reg_idx(struct ixgbe_adapter *adapter, u8 tc,
 
 	switch (hw->mac.type) {
 	case ixgbe_mac_82598EB:
-		*tx = tc << 2;
-		*rx = tc << 3;
+		/* TxQs/TC: 4	RxQs/TC: 8 */
+		*tx = tc << 2; /* 0, 4,  8, 12, 16, 20, 24, 28 */
+		*rx = tc << 3; /* 0, 8, 16, 24, 32, 40, 48, 56 */
 		break;
 	case ixgbe_mac_82599EB:
 	case ixgbe_mac_X540:
 		if (num_tcs > 4) {
-			if (tc < 3) {
-				*tx = tc << 5;
-				*rx = tc << 4;
-			} else if (tc <  5) {
-				*tx = ((tc + 2) << 4);
-				*rx = tc << 4;
-			} else if (tc < num_tcs) {
-				*tx = ((tc + 8) << 3);
-				*rx = tc << 4;
-			}
+			/*
+			 * TCs    : TC0/1 TC2/3 TC4-7
+			 * TxQs/TC:    32    16     8
+			 * RxQs/TC:    16    16    16
+			 */
+			*rx = tc << 4;
+			if (tc < 3)
+				*tx = tc << 5;		/*   0,  32,  64 */
+			else if (tc < 5)
+				*tx = (tc + 2) << 4;	/*  80,  96 */
+			else
+				*tx = (tc + 8) << 3;	/* 104, 112, 120 */
 		} else {
-			*rx =  tc << 5;
-			switch (tc) {
-			case 0:
-				*tx =  0;
-				break;
-			case 1:
-				*tx = 64;
-				break;
-			case 2:
-				*tx = 96;
-				break;
-			case 3:
-				*tx = 112;
-				break;
-			default:
-				break;
-			}
+			/*
+			 * TCs    : TC0 TC1 TC2/3
+			 * TxQs/TC:  64  32    16
+			 * RxQs/TC:  32  32    32
+			 */
+			*rx = tc << 5;
+			if (tc < 2)
+				*tx = tc << 6;		/*  0,  64 */
+			else
+				*tx = (tc + 4) << 4;	/* 96, 112 */
 		}
-		break;
 	default:
 		break;
 	}
@@ -90,25 +85,26 @@ static void ixgbe_get_first_reg_idx(struct ixgbe_adapter *adapter, u8 tc,
  * Cache the descriptor ring offsets for DCB to the assigned rings.
  *
  **/
-static inline bool ixgbe_cache_ring_dcb(struct ixgbe_adapter *adapter)
+static bool ixgbe_cache_ring_dcb(struct ixgbe_adapter *adapter)
 {
 	struct net_device *dev = adapter->netdev;
-	int i, j, k;
+	unsigned int tx_idx, rx_idx;
+	int tc, offset, rss_i, i;
 	u8 num_tcs = netdev_get_num_tc(dev);
 
-	if (!num_tcs)
+	/* verify we have DCB queueing enabled before proceeding */
+	if (num_tcs <= 1)
 		return false;
 
-	for (i = 0, k = 0; i < num_tcs; i++) {
-		unsigned int tx_s, rx_s;
-		u16 count = dev->tc_to_txq[i].count;
+	rss_i = adapter->ring_feature[RING_F_RSS].indices;
 
-		ixgbe_get_first_reg_idx(adapter, i, &tx_s, &rx_s);
-		for (j = 0; j < count; j++, k++) {
-			adapter->tx_ring[k]->reg_idx = tx_s + j;
-			adapter->rx_ring[k]->reg_idx = rx_s + j;
-			adapter->tx_ring[k]->dcb_tc = i;
-			adapter->rx_ring[k]->dcb_tc = i;
+	for (tc = 0, offset = 0; tc < num_tcs; tc++, offset += rss_i) {
+		ixgbe_get_first_reg_idx(adapter, tc, &tx_idx, &rx_idx);
+		for (i = 0; i < rss_i; i++, tx_idx++, rx_idx++) {
+			adapter->tx_ring[offset + i]->reg_idx = tx_idx;
+			adapter->rx_ring[offset + i]->reg_idx = rx_idx;
+			adapter->tx_ring[offset + i]->dcb_tc = tc;
+			adapter->rx_ring[offset + i]->dcb_tc = tc;
 		}
 	}
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 7f2aa22..32c8cd6 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -3608,20 +3608,16 @@ static void ixgbe_configure_dcb(struct ixgbe_adapter *adapter)
 
 	/* Enable RSS Hash per TC */
 	if (hw->mac.type != ixgbe_mac_82598EB) {
-		int i;
-		u32 reg = 0;
-		u8 msb = 0;
-		u8 rss_i = adapter->netdev->tc_to_txq[0].count - 1;
+		u32 msb = 0;
+		u16 rss_i = adapter->ring_feature[RING_F_RSS].indices - 1;
 
 		while (rss_i) {
 			msb++;
 			rss_i >>= 1;
 		}
 
-		for (i = 0; i < MAX_TRAFFIC_CLASS; i++)
-			reg |= msb << IXGBE_RQTC_SHIFT_TC(i);
-
-		IXGBE_WRITE_REG(hw, IXGBE_RQTC, reg);
+		/* write msb to all 8 TCs in one write */
+		IXGBE_WRITE_REG(hw, IXGBE_RQTC, msb * 0x11111111);
 	}
 }
 #endif
-- 
1.7.10.4

[net-next 5/6] ixgbe: Cleanup logic for MRQC and MTQC configuration

[Jeff Kirsher]

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

This change is meant to make the code much more readable for MTQC and MRQC
configuration.

The big change is that I simplified much of the logic so that we are
essentially handling just 4 cases and their variants. In the cases where
RSS is disabled we are actually just programming the RETA table with all
1s resulting in a single queue RSS. In the case of SR-IOV I am treating
that as a subset of VMDq. This all results int he following configuration
for the hardware:
         DCB
         En       Dis
VMDq En  VMDQ/DCB VMDq/RSS
     Dis DCB/RSS  RSS

Cc: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Stephen Ko <stephen.s.ko@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |  116 ++++++++++++++-----------
 1 file changed, 66 insertions(+), 50 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 32c8cd6..2b4b791 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -2719,8 +2719,7 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter,
 static void ixgbe_setup_mtqc(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	u32 rttdcs;
-	u32 reg;
+	u32 rttdcs, mtqc;
 	u8 tcs = netdev_get_num_tc(adapter->netdev);
 
 	if (hw->mac.type == ixgbe_mac_82598EB)
@@ -2732,28 +2731,32 @@ static void ixgbe_setup_mtqc(struct ixgbe_adapter *adapter)
 	IXGBE_WRITE_REG(hw, IXGBE_RTTDCS, rttdcs);
 
 	/* set transmit pool layout */
-	switch (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
-	case (IXGBE_FLAG_SRIOV_ENABLED):
-		IXGBE_WRITE_REG(hw, IXGBE_MTQC,
-				(IXGBE_MTQC_VT_ENA | IXGBE_MTQC_64VF));
-		break;
-	default:
-		if (!tcs)
-			reg = IXGBE_MTQC_64Q_1PB;
-		else if (tcs <= 4)
-			reg = IXGBE_MTQC_RT_ENA | IXGBE_MTQC_4TC_4TQ;
+	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
+		mtqc = IXGBE_MTQC_VT_ENA;
+		if (tcs > 4)
+			mtqc |= IXGBE_MTQC_RT_ENA | IXGBE_MTQC_8TC_8TQ;
+		else if (tcs > 1)
+			mtqc |= IXGBE_MTQC_RT_ENA | IXGBE_MTQC_4TC_4TQ;
+		else if (adapter->ring_feature[RING_F_RSS].indices == 4)
+			mtqc |= IXGBE_MTQC_32VF;
 		else
-			reg = IXGBE_MTQC_RT_ENA | IXGBE_MTQC_8TC_8TQ;
+			mtqc |= IXGBE_MTQC_64VF;
+	} else {
+		if (tcs > 4)
+			mtqc = IXGBE_MTQC_RT_ENA | IXGBE_MTQC_8TC_8TQ;
+		else if (tcs > 1)
+			mtqc = IXGBE_MTQC_RT_ENA | IXGBE_MTQC_4TC_4TQ;
+		else
+			mtqc = IXGBE_MTQC_64Q_1PB;
+	}
 
-		IXGBE_WRITE_REG(hw, IXGBE_MTQC, reg);
+	IXGBE_WRITE_REG(hw, IXGBE_MTQC, mtqc);
 
-		/* Enable Security TX Buffer IFG for multiple pb */
-		if (tcs) {
-			reg = IXGBE_READ_REG(hw, IXGBE_SECTXMINIFG);
-			reg |= IXGBE_SECTX_DCB;
-			IXGBE_WRITE_REG(hw, IXGBE_SECTXMINIFG, reg);
-		}
-		break;
+	/* Enable Security TX Buffer IFG for multiple pb */
+	if (tcs) {
+		u32 sectx = IXGBE_READ_REG(hw, IXGBE_SECTXMINIFG);
+		sectx |= IXGBE_SECTX_DCB;
+		IXGBE_WRITE_REG(hw, IXGBE_SECTXMINIFG, sectx);
 	}
 
 	/* re-enable the arbiter */
@@ -2886,11 +2889,18 @@ static void ixgbe_setup_mrqc(struct ixgbe_adapter *adapter)
 	u32 mrqc = 0, reta = 0;
 	u32 rxcsum;
 	int i, j;
-	u8 tcs = netdev_get_num_tc(adapter->netdev);
-	int maxq = adapter->ring_feature[RING_F_RSS].indices;
+	u16 rss_i = adapter->ring_feature[RING_F_RSS].indices;
+
+	if (!(adapter->flags & IXGBE_FLAG_RSS_ENABLED))
+		rss_i = 1;
 
-	if (tcs)
-		maxq = min(maxq, adapter->num_tx_queues / tcs);
+	/*
+	 * Program table for at least 2 queues w/ SR-IOV so that VFs can
+	 * make full use of any rings they may have.  We will use the
+	 * PSRTYPE register to control how many rings we use within the PF.
+	 */
+	if ((adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) && (rss_i < 2))
+		rss_i = 2;
 
 	/* Fill out hash function seeds */
 	for (i = 0; i < 10; i++)
@@ -2898,7 +2908,7 @@ static void ixgbe_setup_mrqc(struct ixgbe_adapter *adapter)
 
 	/* Fill out redirection table */
 	for (i = 0, j = 0; i < 128; i++, j++) {
-		if (j == maxq)
+		if (j == rss_i)
 			j = 0;
 		/* reta = 4-byte sliding window of
 		 * 0x00..(indices-1)(indices-1)00..etc. */
@@ -2912,35 +2922,36 @@ static void ixgbe_setup_mrqc(struct ixgbe_adapter *adapter)
 	rxcsum |= IXGBE_RXCSUM_PCSD;
 	IXGBE_WRITE_REG(hw, IXGBE_RXCSUM, rxcsum);
 
-	if (adapter->hw.mac.type == ixgbe_mac_82598EB &&
-	    (adapter->flags & IXGBE_FLAG_RSS_ENABLED)) {
-		mrqc = IXGBE_MRQC_RSSEN;
+	if (adapter->hw.mac.type == ixgbe_mac_82598EB) {
+		if (adapter->flags & IXGBE_FLAG_RSS_ENABLED)
+			mrqc = IXGBE_MRQC_RSSEN;
 	} else {
-		int mask = adapter->flags & (IXGBE_FLAG_RSS_ENABLED
-					     | IXGBE_FLAG_SRIOV_ENABLED);
-
-		switch (mask) {
-		case (IXGBE_FLAG_RSS_ENABLED):
-			if (!tcs)
-				mrqc = IXGBE_MRQC_RSSEN;
-			else if (tcs <= 4)
-				mrqc = IXGBE_MRQC_RTRSS4TCEN;
+		u8 tcs = netdev_get_num_tc(adapter->netdev);
+
+		if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
+			if (tcs > 4)
+				mrqc = IXGBE_MRQC_VMDQRT8TCEN;	/* 8 TCs */
+			else if (tcs > 1)
+				mrqc = IXGBE_MRQC_VMDQRT4TCEN;	/* 4 TCs */
+			else if (adapter->ring_feature[RING_F_RSS].indices == 4)
+				mrqc = IXGBE_MRQC_VMDQRSS32EN;
 			else
+				mrqc = IXGBE_MRQC_VMDQRSS64EN;
+		} else {
+			if (tcs > 4)
 				mrqc = IXGBE_MRQC_RTRSS8TCEN;
-			break;
-		case (IXGBE_FLAG_SRIOV_ENABLED):
-			mrqc = IXGBE_MRQC_VMDQEN;
-			break;
-		default:
-			break;
+			else if (tcs > 1)
+				mrqc = IXGBE_MRQC_RTRSS4TCEN;
+			else
+				mrqc = IXGBE_MRQC_RSSEN;
 		}
 	}
 
 	/* Perform hash on these packet types */
-	mrqc |= IXGBE_MRQC_RSS_FIELD_IPV4
-	      | IXGBE_MRQC_RSS_FIELD_IPV4_TCP
-	      | IXGBE_MRQC_RSS_FIELD_IPV6
-	      | IXGBE_MRQC_RSS_FIELD_IPV6_TCP;
+	mrqc |= IXGBE_MRQC_RSS_FIELD_IPV4 |
+		IXGBE_MRQC_RSS_FIELD_IPV4_TCP |
+		IXGBE_MRQC_RSS_FIELD_IPV6 |
+		IXGBE_MRQC_RSS_FIELD_IPV6_TCP;
 
 	if (adapter->flags2 & IXGBE_FLAG2_RSS_FIELD_IPV4_UDP)
 		mrqc |= IXGBE_MRQC_RSS_FIELD_IPV4_UDP;
@@ -3103,8 +3114,13 @@ static void ixgbe_setup_psrtype(struct ixgbe_adapter *adapter)
 	if (hw->mac.type == ixgbe_mac_82598EB)
 		return;
 
-	if (adapter->flags & IXGBE_FLAG_RSS_ENABLED)
-		psrtype |= (adapter->num_rx_queues_per_pool << 29);
+	if (adapter->flags & IXGBE_FLAG_RSS_ENABLED) {
+		int rss_i = adapter->ring_feature[RING_F_RSS].indices;
+		if (rss_i > 3)
+			psrtype |= 2 << 29;
+		else if (rss_i > 1)
+			psrtype |= 1 << 29;
+	}
 
 	for (p = 0; p < adapter->num_rx_pools; p++)
 		IXGBE_WRITE_REG(hw, IXGBE_PSRTYPE(adapter->num_vfs + p),
-- 
1.7.10.4

[net-next 6/6] ixgbevf: Update descriptor macros to accept pointers and drop _ADV suffix

[Jeff Kirsher]

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

This change updates the descriptor macros to accept pointers, updates the
name to drop the _ADV suffix, and include the IXGBEVF name in the macro.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Sibai Li <sibai.li@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/ixgbevf/ixgbevf.h      |   12 ++++++------
 drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c |   18 +++++++++---------
 2 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
index f92daca..1f13765 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
@@ -164,12 +164,12 @@ struct ixgbevf_q_vector {
 	((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \
 	(R)->next_to_clean - (R)->next_to_use - 1)
 
-#define IXGBE_RX_DESC_ADV(R, i)	    \
-	(&(((union ixgbe_adv_rx_desc *)((R).desc))[i]))
-#define IXGBE_TX_DESC_ADV(R, i)	    \
-	(&(((union ixgbe_adv_tx_desc *)((R).desc))[i]))
-#define IXGBE_TX_CTXTDESC_ADV(R, i)	    \
-	(&(((struct ixgbe_adv_tx_context_desc *)((R).desc))[i]))
+#define IXGBEVF_RX_DESC(R, i)	    \
+	(&(((union ixgbe_adv_rx_desc *)((R)->desc))[i]))
+#define IXGBEVF_TX_DESC(R, i)	    \
+	(&(((union ixgbe_adv_tx_desc *)((R)->desc))[i]))
+#define IXGBEVF_TX_CTXTDESC(R, i)	    \
+	(&(((struct ixgbe_adv_tx_context_desc *)((R)->desc))[i]))
 
 #define IXGBE_MAX_JUMBO_FRAME_SIZE        16128
 
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index 8e022c6..c98cdf7 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -195,7 +195,7 @@ static bool ixgbevf_clean_tx_irq(struct ixgbevf_q_vector *q_vector,
 
 	i = tx_ring->next_to_clean;
 	eop = tx_ring->tx_buffer_info[i].next_to_watch;
-	eop_desc = IXGBE_TX_DESC_ADV(*tx_ring, eop);
+	eop_desc = IXGBEVF_TX_DESC(tx_ring, eop);
 
 	while ((eop_desc->wb.status & cpu_to_le32(IXGBE_TXD_STAT_DD)) &&
 	       (count < tx_ring->count)) {
@@ -206,7 +206,7 @@ static bool ixgbevf_clean_tx_irq(struct ixgbevf_q_vector *q_vector,
 			goto cont_loop;
 		for ( ; !cleaned; count++) {
 			struct sk_buff *skb;
-			tx_desc = IXGBE_TX_DESC_ADV(*tx_ring, i);
+			tx_desc = IXGBEVF_TX_DESC(tx_ring, i);
 			tx_buffer_info = &tx_ring->tx_buffer_info[i];
 			cleaned = (i == eop);
 			skb = tx_buffer_info->skb;
@@ -235,7 +235,7 @@ static bool ixgbevf_clean_tx_irq(struct ixgbevf_q_vector *q_vector,
 
 cont_loop:
 		eop = tx_ring->tx_buffer_info[i].next_to_watch;
-		eop_desc = IXGBE_TX_DESC_ADV(*tx_ring, eop);
+		eop_desc = IXGBEVF_TX_DESC(tx_ring, eop);
 	}
 
 	tx_ring->next_to_clean = i;
@@ -339,7 +339,7 @@ static void ixgbevf_alloc_rx_buffers(struct ixgbevf_adapter *adapter,
 	bi = &rx_ring->rx_buffer_info[i];
 
 	while (cleaned_count--) {
-		rx_desc = IXGBE_RX_DESC_ADV(*rx_ring, i);
+		rx_desc = IXGBEVF_RX_DESC(rx_ring, i);
 		skb = bi->skb;
 		if (!skb) {
 			skb = netdev_alloc_skb(adapter->netdev,
@@ -405,7 +405,7 @@ static bool ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector,
 	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
 
 	i = rx_ring->next_to_clean;
-	rx_desc = IXGBE_RX_DESC_ADV(*rx_ring, i);
+	rx_desc = IXGBEVF_RX_DESC(rx_ring, i);
 	staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
 	rx_buffer_info = &rx_ring->rx_buffer_info[i];
 
@@ -432,7 +432,7 @@ static bool ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector,
 		if (i == rx_ring->count)
 			i = 0;
 
-		next_rxd = IXGBE_RX_DESC_ADV(*rx_ring, i);
+		next_rxd = IXGBEVF_RX_DESC(rx_ring, i);
 		prefetch(next_rxd);
 		cleaned_count++;
 
@@ -2437,7 +2437,7 @@ static int ixgbevf_tso(struct ixgbevf_adapter *adapter,
 		i = tx_ring->next_to_use;
 
 		tx_buffer_info = &tx_ring->tx_buffer_info[i];
-		context_desc = IXGBE_TX_CTXTDESC_ADV(*tx_ring, i);
+		context_desc = IXGBEVF_TX_CTXTDESC(tx_ring, i);
 
 		/* VLAN MACLEN IPLEN */
 		if (tx_flags & IXGBE_TX_FLAGS_VLAN)
@@ -2497,7 +2497,7 @@ static bool ixgbevf_tx_csum(struct ixgbevf_adapter *adapter,
 	    (tx_flags & IXGBE_TX_FLAGS_VLAN)) {
 		i = tx_ring->next_to_use;
 		tx_buffer_info = &tx_ring->tx_buffer_info[i];
-		context_desc = IXGBE_TX_CTXTDESC_ADV(*tx_ring, i);
+		context_desc = IXGBEVF_TX_CTXTDESC(tx_ring, i);
 
 		if (tx_flags & IXGBE_TX_FLAGS_VLAN)
 			vlan_macip_lens |= (tx_flags &
@@ -2700,7 +2700,7 @@ static void ixgbevf_tx_queue(struct ixgbevf_adapter *adapter,
 	i = tx_ring->next_to_use;
 	while (count--) {
 		tx_buffer_info = &tx_ring->tx_buffer_info[i];
-		tx_desc = IXGBE_TX_DESC_ADV(*tx_ring, i);
+		tx_desc = IXGBEVF_TX_DESC(tx_ring, i);
 		tx_desc->read.buffer_addr = cpu_to_le64(tx_buffer_info->dma);
 		tx_desc->read.cmd_type_len =
 			cpu_to_le32(cmd_type_len | tx_buffer_info->length);
-- 
1.7.10.4

Re: [net-next 0/6][pull request] Intel Wired LAN Driver Updates

[David Miller]

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Tue, 17 Jul 2012 19:20:39 -0700

> This series contains updates to ixgbe & ixgbevf.
> 
> The following are changes since commit 5abf7f7e0f6bdbfcac737f636497d7016d9507eb:
>   ipv4: fix rcu splat
> and are available in the git repository at:
>   git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master

Pulled, thanks Jeff.