[net-next-2.6 0/3][pull request] Intel Wired LAN Driver Updates

[net-next-2.6 0/3][pull request] Intel Wired LAN Driver Updates

[Jeff Kirsher]

The following series contains the addition of ixgbe ethtool support
for 100MB FULL and the addition of igb PF support for i350 devices as
well as VF transmit rate limit using iproute2.

The following are changes since commit a4daad6b0923030fbd3b00a01f570e4c3eef446b:
  net: Pre-COW metrics for TCP.

and are available in the git repository at:
  master.kernel.org:/pub/scm/linux/kernel/git/jkirsher/net-next-2.6 master

Atita Shirwaikar (1):
  ixgbe: Adding 100MB FULL support in ethtool

Carolyn Wyborny (1):
  igb: Enable PF side of SR-IOV support for i350 devices

Lior Levy (1):
  igb: add support for VF Transmit rate limit using iproute2

 drivers/net/igb/e1000_82575.c     |   10 +++-
 drivers/net/igb/e1000_defines.h   |    7 +++
 drivers/net/igb/e1000_mbx.c       |   38 ++++++-------
 drivers/net/igb/e1000_regs.h      |    4 ++
 drivers/net/igb/igb.h             |    2 +
 drivers/net/igb/igb_main.c        |  105 +++++++++++++++++++++++++++++++++++--
 drivers/net/ixgbe/ixgbe_ethtool.c |   34 +++++++++++-
 drivers/net/ixgbe/ixgbe_main.c    |    5 ++-
 8 files changed, 176 insertions(+), 29 deletions(-)

-- 
1.7.3.5

[net-next-2.6 1/3] igb: Enable PF side of SR-IOV support for i350 devices

[Jeff Kirsher]

From: Carolyn Wyborny <carolyn.wyborny@intel.com>

This patch adds full support for SR-IOV by enabling the PF side.
VF side has already been committed.

Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/igb/e1000_82575.c |   10 ++++++++--
 drivers/net/igb/e1000_mbx.c   |   38 ++++++++++++++++++--------------------
 drivers/net/igb/igb_main.c    |    9 +++++++--
 3 files changed, 33 insertions(+), 24 deletions(-)

diff --git a/drivers/net/igb/e1000_82575.c b/drivers/net/igb/e1000_82575.c
index c1552b6..65c1833 100644
--- a/drivers/net/igb/e1000_82575.c
+++ b/drivers/net/igb/e1000_82575.c
@@ -238,9 +238,15 @@ static s32 igb_get_invariants_82575(struct e1000_hw *hw)
 		size = 14;
 	nvm->word_size = 1 << size;
 
-	/* if 82576 then initialize mailbox parameters */
-	if (mac->type == e1000_82576)
+	/* if part supports SR-IOV then initialize mailbox parameters */
+	switch (mac->type) {
+	case e1000_82576:
+	case e1000_i350:
 		igb_init_mbx_params_pf(hw);
+		break;
+	default:
+		break;
+	}
 
 	/* setup PHY parameters */
 	if (phy->media_type != e1000_media_type_copper) {
diff --git a/drivers/net/igb/e1000_mbx.c b/drivers/net/igb/e1000_mbx.c
index c474cdb..78d48c7 100644
--- a/drivers/net/igb/e1000_mbx.c
+++ b/drivers/net/igb/e1000_mbx.c
@@ -422,26 +422,24 @@ s32 igb_init_mbx_params_pf(struct e1000_hw *hw)
 {
 	struct e1000_mbx_info *mbx = &hw->mbx;
 
-	if (hw->mac.type == e1000_82576) {
-		mbx->timeout = 0;
-		mbx->usec_delay = 0;
-
-		mbx->size = E1000_VFMAILBOX_SIZE;
-
-		mbx->ops.read = igb_read_mbx_pf;
-		mbx->ops.write = igb_write_mbx_pf;
-		mbx->ops.read_posted = igb_read_posted_mbx;
-		mbx->ops.write_posted = igb_write_posted_mbx;
-		mbx->ops.check_for_msg = igb_check_for_msg_pf;
-		mbx->ops.check_for_ack = igb_check_for_ack_pf;
-		mbx->ops.check_for_rst = igb_check_for_rst_pf;
-
-		mbx->stats.msgs_tx = 0;
-		mbx->stats.msgs_rx = 0;
-		mbx->stats.reqs = 0;
-		mbx->stats.acks = 0;
-		mbx->stats.rsts = 0;
-	}
+	mbx->timeout = 0;
+	mbx->usec_delay = 0;
+
+	mbx->size = E1000_VFMAILBOX_SIZE;
+
+	mbx->ops.read = igb_read_mbx_pf;
+	mbx->ops.write = igb_write_mbx_pf;
+	mbx->ops.read_posted = igb_read_posted_mbx;
+	mbx->ops.write_posted = igb_write_posted_mbx;
+	mbx->ops.check_for_msg = igb_check_for_msg_pf;
+	mbx->ops.check_for_ack = igb_check_for_ack_pf;
+	mbx->ops.check_for_rst = igb_check_for_rst_pf;
+
+	mbx->stats.msgs_tx = 0;
+	mbx->stats.msgs_rx = 0;
+	mbx->stats.reqs = 0;
+	mbx->stats.acks = 0;
+	mbx->stats.rsts = 0;
 
 	return 0;
 }
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 200cc32..cb6bf7b 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -2287,9 +2287,14 @@ static int __devinit igb_sw_init(struct igb_adapter *adapter)
 
 	spin_lock_init(&adapter->stats64_lock);
 #ifdef CONFIG_PCI_IOV
-	if (hw->mac.type == e1000_82576)
+	switch (hw->mac.type) {
+	case e1000_82576:
+	case e1000_i350:
 		adapter->vfs_allocated_count = (max_vfs > 7) ? 7 : max_vfs;
-
+		break;
+	default:
+		break;
+	}
 #endif /* CONFIG_PCI_IOV */
 	adapter->rss_queues = min_t(u32, IGB_MAX_RX_QUEUES, num_online_cpus());
 
-- 
1.7.3.5

[net-next-2.6 2/3] igb: add support for VF Transmit rate limit using iproute2

[Jeff Kirsher]

From: Lior Levy <lior.levy@intel.com>

Implemented igb_ndo_set_vf_bw function which is being used
by iproute2 tool. In addition, updated igb_ndo_get_vf_config function
to show the actual rate limit to the user.

The rate limitation can be configured only when the link is up.
The rate limit value can be ranged between 0 and actual
link speed measured in Mbps. A value of '0' disables the rate limit for
this specific VF.

iproute2 usage will be 'ip link set ethX vf Y rate Z'.
After the command is made, the rate will be changed instantly.
To view the current rate limit, use 'ip link show ethX'.

The rates will be zeroed only upon driver reload or a link speed change.

This feature is being supported only by 82576 device.

Signed-off-by: Lior Levy <lior.levy@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/igb/e1000_defines.h |    7 +++
 drivers/net/igb/e1000_regs.h    |    4 ++
 drivers/net/igb/igb.h           |    2 +
 drivers/net/igb/igb_main.c      |   96 ++++++++++++++++++++++++++++++++++++++-
 4 files changed, 107 insertions(+), 2 deletions(-)

diff --git a/drivers/net/igb/e1000_defines.h b/drivers/net/igb/e1000_defines.h
index 6319ed9..0bd543f 100644
--- a/drivers/net/igb/e1000_defines.h
+++ b/drivers/net/igb/e1000_defines.h
@@ -770,4 +770,11 @@
 #define E1000_PCIEMISC_LX_DECISION      0x00000080 /* Lx power decision based
                                                       on DMA coal */
 
+/* Tx Rate-Scheduler Config fields */
+#define E1000_RTTBCNRC_RS_ENA          0x80000000
+#define E1000_RTTBCNRC_RF_DEC_MASK     0x00003FFF
+#define E1000_RTTBCNRC_RF_INT_SHIFT    14
+#define E1000_RTTBCNRC_RF_INT_MASK     \
+	(E1000_RTTBCNRC_RF_DEC_MASK << E1000_RTTBCNRC_RF_INT_SHIFT)
+
 #endif
diff --git a/drivers/net/igb/e1000_regs.h b/drivers/net/igb/e1000_regs.h
index 8ac83c5..a6485a1 100644
--- a/drivers/net/igb/e1000_regs.h
+++ b/drivers/net/igb/e1000_regs.h
@@ -106,6 +106,10 @@
 
 #define E1000_RQDPC(_n) (0x0C030 + ((_n) * 0x40))
 
+/* TX Rate Limit Registers */
+#define E1000_RTTDQSEL 0x3604  /* Tx Desc Plane Queue Select - WO */
+#define E1000_RTTBCNRC 0x36B0  /* Tx BCN Rate-Scheduler Config - WO */
+
 /* Split and Replication RX Control - RW */
 #define E1000_RXPBS    0x02404  /* Rx Packet Buffer Size - RW */
 /*
diff --git a/drivers/net/igb/igb.h b/drivers/net/igb/igb.h
index 92a4ef0..bbc5ebf 100644
--- a/drivers/net/igb/igb.h
+++ b/drivers/net/igb/igb.h
@@ -77,6 +77,7 @@ struct vf_data_storage {
 	unsigned long last_nack;
 	u16 pf_vlan; /* When set, guest VLAN config not allowed. */
 	u16 pf_qos;
+	u16 tx_rate;
 };
 
 #define IGB_VF_FLAG_CTS            0x00000001 /* VF is clear to send data */
@@ -323,6 +324,7 @@ struct igb_adapter {
 	u16 rx_ring_count;
 	unsigned int vfs_allocated_count;
 	struct vf_data_storage *vf_data;
+	int vf_rate_link_speed;
 	u32 rss_queues;
 	u32 wvbr;
 };
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index cb6bf7b..6b17317 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -150,6 +150,7 @@ static int igb_ndo_set_vf_vlan(struct net_device *netdev,
 static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate);
 static int igb_ndo_get_vf_config(struct net_device *netdev, int vf,
 				 struct ifla_vf_info *ivi);
+static void igb_check_vf_rate_limit(struct igb_adapter *);
 
 #ifdef CONFIG_PM
 static int igb_suspend(struct pci_dev *, pm_message_t);
@@ -3511,6 +3512,7 @@ static void igb_watchdog_task(struct work_struct *work)
 			netif_carrier_on(netdev);
 
 			igb_ping_all_vfs(adapter);
+			igb_check_vf_rate_limit(adapter);
 
 			/* link state has changed, schedule phy info update */
 			if (!test_bit(__IGB_DOWN, &adapter->state))
@@ -6599,9 +6601,99 @@ static int igb_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
 	return igb_set_vf_mac(adapter, vf, mac);
 }
 
+static int igb_link_mbps(int internal_link_speed)
+{
+	switch (internal_link_speed) {
+	case SPEED_100:
+		return 100;
+	case SPEED_1000:
+		return 1000;
+	default:
+		return 0;
+	}
+}
+
+static void igb_set_vf_rate_limit(struct e1000_hw *hw, int vf, int tx_rate,
+                                  int link_speed)
+{
+	int rf_dec, rf_int;
+	u32 bcnrc_val;
+
+	if (tx_rate != 0) {
+		/* Calculate the rate factor values to set */
+		rf_int = link_speed / tx_rate;
+		rf_dec = (link_speed - (rf_int * tx_rate));
+		rf_dec = (rf_dec * (1 << E1000_RTTBCNRC_RF_INT_SHIFT)) /
+		         tx_rate;
+
+		bcnrc_val = E1000_RTTBCNRC_RS_ENA;
+		bcnrc_val |= ((rf_int << E1000_RTTBCNRC_RF_INT_SHIFT) &
+		               E1000_RTTBCNRC_RF_INT_MASK);
+		bcnrc_val |= (rf_dec & E1000_RTTBCNRC_RF_DEC_MASK);
+	} else
+		bcnrc_val = 0;
+
+	wr32(E1000_RTTDQSEL, vf); /* vf X uses queue X */
+	wr32(E1000_RTTBCNRC, bcnrc_val);
+}
+
+static void igb_check_vf_rate_limit(struct igb_adapter *adapter)
+{
+	int actual_link_speed, i;
+	bool reset_rate = false;
+
+	/* VF TX rate limit was not set or not supported */
+	if ((adapter->vf_rate_link_speed == 0) ||
+	    (adapter->hw.mac.type != e1000_82576))
+		return;
+
+	actual_link_speed = igb_link_mbps(adapter->link_speed);
+	if (actual_link_speed != adapter->vf_rate_link_speed) {
+		reset_rate = true;
+		adapter->vf_rate_link_speed = 0;
+		dev_info(&adapter->pdev->dev,
+		         "Link speed has been changed. VF Transmit "
+		         "rate is disabled\n");
+	}
+
+	for (i = 0; i < adapter->vfs_allocated_count; i++) {
+		if (reset_rate)
+			adapter->vf_data[i].tx_rate = 0;
+
+		igb_set_vf_rate_limit(&adapter->hw, i,
+		                      adapter->vf_data[i].tx_rate,
+		                      actual_link_speed);
+	}
+}
+
 static int igb_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate)
 {
-	return -EOPNOTSUPP;
+	struct igb_adapter *adapter = netdev_priv(netdev);
+	struct e1000_hw *hw = &adapter->hw;
+	int actual_link_speed;
+
+	if (hw->mac.type != e1000_82576)
+		return -EOPNOTSUPP;
+
+	actual_link_speed = igb_link_mbps(adapter->link_speed);
+	if ((vf >= adapter->vfs_allocated_count) ||
+	    (!(rd32(E1000_STATUS) & E1000_STATUS_LU)) ||
+	    (tx_rate < 0) || (tx_rate > actual_link_speed))
+		return -EINVAL;
+
+	adapter->vf_rate_link_speed = actual_link_speed;
+	adapter->vf_data[vf].tx_rate = (u16)tx_rate;
+	igb_set_vf_rate_limit(hw, vf, tx_rate, actual_link_speed);
+
+	if (tx_rate != 0)
+		dev_info(&adapter->pdev->dev,
+		         "Setting Transmit rate of %d Mbps for VF %d\n",
+		         tx_rate, vf);
+	else
+		dev_info(&adapter->pdev->dev,
+		         "Transmit rate limit for VF %d is disabled\n", vf);
+
+	return 0;
 }
 
 static int igb_ndo_get_vf_config(struct net_device *netdev,
@@ -6612,7 +6704,7 @@ static int igb_ndo_get_vf_config(struct net_device *netdev,
 		return -EINVAL;
 	ivi->vf = vf;
 	memcpy(&ivi->mac, adapter->vf_data[vf].vf_mac_addresses, ETH_ALEN);
-	ivi->tx_rate = 0;
+	ivi->tx_rate = adapter->vf_data[vf].tx_rate;
 	ivi->vlan = adapter->vf_data[vf].pf_vlan;
 	ivi->qos = adapter->vf_data[vf].pf_qos;
 	return 0;
-- 
1.7.3.5

[net-next-2.6 3/3] ixgbe: Adding 100MB FULL support in ethtool

[Jeff Kirsher]

From: Atita Shirwaikar <atita.shirwaikar@intel.com>

Current driver does not show 100MB support in ethtool.
Adding support for the same.

Signed-off-by: Atita Shirwaikar <atita.shirwaikar@intel.com>
Tested-by: Stephen Ko <stephen.s.ko@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ixgbe/ixgbe_ethtool.c |   34 ++++++++++++++++++++++++++++++++--
 drivers/net/ixgbe/ixgbe_main.c    |    5 ++++-
 2 files changed, 36 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index 2002ea8..309272f 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -152,7 +152,17 @@ static int ixgbe_get_settings(struct net_device *netdev,
 		ecmd->supported |= (SUPPORTED_1000baseT_Full |
 		                    SUPPORTED_Autoneg);
 
+		switch (hw->mac.type) {
+		case ixgbe_mac_X540:
+			ecmd->supported |= SUPPORTED_100baseT_Full;
+			break;
+		default:
+			break;
+		}
+
 		ecmd->advertising = ADVERTISED_Autoneg;
+		if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_100_FULL)
+			ecmd->advertising |= ADVERTISED_100baseT_Full;
 		if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_10GB_FULL)
 			ecmd->advertising |= ADVERTISED_10000baseT_Full;
 		if (hw->phy.autoneg_advertised & IXGBE_LINK_SPEED_1GB_FULL)
@@ -167,6 +177,15 @@ static int ixgbe_get_settings(struct net_device *netdev,
 			ecmd->advertising |= (ADVERTISED_10000baseT_Full |
 					      ADVERTISED_1000baseT_Full);
 
+		switch (hw->mac.type) {
+		case ixgbe_mac_X540:
+			if (!(ecmd->advertising & ADVERTISED_100baseT_Full))
+				ecmd->advertising |= (ADVERTISED_100baseT_Full);
+			break;
+		default:
+			break;
+		}
+
 		if (hw->phy.media_type == ixgbe_media_type_copper) {
 			ecmd->supported |= SUPPORTED_TP;
 			ecmd->advertising |= ADVERTISED_TP;
@@ -271,8 +290,19 @@ static int ixgbe_get_settings(struct net_device *netdev,
 
 	hw->mac.ops.check_link(hw, &link_speed, &link_up, false);
 	if (link_up) {
-		ecmd->speed = (link_speed == IXGBE_LINK_SPEED_10GB_FULL) ?
-		               SPEED_10000 : SPEED_1000;
+		switch (link_speed) {
+		case IXGBE_LINK_SPEED_10GB_FULL:
+			ecmd->speed = SPEED_10000;
+			break;
+		case IXGBE_LINK_SPEED_1GB_FULL:
+			ecmd->speed = SPEED_1000;
+			break;
+		case IXGBE_LINK_SPEED_100_FULL:
+			ecmd->speed = SPEED_100;
+			break;
+		default:
+			break;
+		}
 		ecmd->duplex = DUPLEX_FULL;
 	} else {
 		ecmd->speed = -1;
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 602078b..b923e42 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -6095,7 +6095,10 @@ static void ixgbe_watchdog_task(struct work_struct *work)
 			       (link_speed == IXGBE_LINK_SPEED_10GB_FULL ?
 			       "10 Gbps" :
 			       (link_speed == IXGBE_LINK_SPEED_1GB_FULL ?
-			       "1 Gbps" : "unknown speed")),
+			       "1 Gbps" :
+			       (link_speed == IXGBE_LINK_SPEED_100_FULL ?
+			       "100 Mbps" :
+			       "unknown speed"))),
 			       ((flow_rx && flow_tx) ? "RX/TX" :
 			       (flow_rx ? "RX" :
 			       (flow_tx ? "TX" : "None"))));
-- 
1.7.3.5

Re: [net-next-2.6 2/3] igb: add support for VF Transmit rate limit using iproute2

[David Miller]

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Fri, 28 Jan 2011 04:29:38 -0800

> +	if (tx_rate != 0)
> +		dev_info(&adapter->pdev->dev,
> +		         "Setting Transmit rate of %d Mbps for VF %d\n",
> +		         tx_rate, vf);
> +	else
> +		dev_info(&adapter->pdev->dev,
> +		         "Transmit rate limit for VF %d is disabled\n", vf);

If you're going to print this, use netdev_info(netdev, ...).

But I think you shouldn't be logging anything at all.

No other ethtool operation logs what it did except in extremely
exceptional error conditions.  And there is nothing special
about this VF rate limiting ethtool operation to justify these
extraneous logging messages.

If people want to know if the VF is rate limited, and by how much,
then can query the configuration using ethtool.

RE: [net-next-2.6 2/3] igb: add support for VF Transmit rate limit using iproute2

[Levy, Lior]

Thank you Dave for your input.

I agree that there isn't a very good reason to log these messages.
People can use the "ip show" command to query the TX rate limit for each vf.

BTW - this feature is being configured via iproute2 and not ethtool. But still the concept is the same.

We'll re-submit the patch soon.
 
Regards,
Lior.

-----Original Message-----
From: netdev-owner@vger.kernel.org [mailto:netdev-owner@vger.kernel.org] On Behalf Of David Miller
Sent: Friday, January 28, 2011 4:38 PM
To: Kirsher, Jeffrey T
Cc: Levy, Lior; netdev@vger.kernel.org; gospo@redhat.com; bphilips@novell.com
Subject: Re: [net-next-2.6 2/3] igb: add support for VF Transmit rate limit using iproute2

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Fri, 28 Jan 2011 04:29:38 -0800

> +	if (tx_rate != 0)
> +		dev_info(&adapter->pdev->dev,
> +		         "Setting Transmit rate of %d Mbps for VF %d\n",
> +		         tx_rate, vf);
> +	else
> +		dev_info(&adapter->pdev->dev,
> +		         "Transmit rate limit for VF %d is disabled\n", vf);

If you're going to print this, use netdev_info(netdev, ...).

But I think you shouldn't be logging anything at all.

No other ethtool operation logs what it did except in extremely
exceptional error conditions.  And there is nothing special
about this VF rate limiting ethtool operation to justify these
extraneous logging messages.

If people want to know if the VF is rate limited, and by how much,
then can query the configuration using ethtool.
--
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