[net-next 0/7][pull request] Intel Wired LAN Driver Updates

9 messages, 2 authors, 2012-07-17 · open the first message on its own page

[net-next 0/7][pull request] Intel Wired LAN Driver Updates

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:25

This series contains updates to e1000e and ixgbe.

The following are changes since commit 141e369de698f2e17bf716b83fcc647ddcb2220c:
  xfrm: Initialize the struct xfrm_dst behind the dst_enty field
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master

Alexander Duyck (5):
  ixgbe: Simplify logic for getting traffic class from user priority
  ixgbe: Cleanup unpacking code for DCB
  ixgbe: Populate the prio_tc_map in ixgbe_setup_tc
  ixgbe: Add function for obtaining FCoE TC based on FCoE user priority
  ixgbe: Merge FCoE set_num and cache_ring calls into RSS/DCB config

Matthew Vick (1):
  e1000e: Program the correct register for ITR when using MSI-X.

Tushar Dave (1):
  e1000e: Cleanup code logic in e1000_check_for_serdes_link_82571()

 drivers/net/ethernet/intel/e1000e/82571.c       |   14 +-
 drivers/net/ethernet/intel/e1000e/e1000.h       |    1 +
 drivers/net/ethernet/intel/e1000e/ethtool.c     |    5 +-
 drivers/net/ethernet/intel/e1000e/netdev.c      |   32 ++-
 drivers/net/ethernet/intel/ixgbe/ixgbe.h        |    1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c    |   74 ++++---
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.h    |    1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c |   26 +--
 drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c   |   15 ++
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c    |  260 ++++++++++-------------
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c   |   61 ++++--
 11 files changed, 265 insertions(+), 225 deletions(-)

-- 
1.7.10.4

[net-next 1/7] e1000e: Cleanup code logic in e1000_check_for_serdes_link_82571()

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:29

From: Tushar Dave <redacted>

Cleanup code to make it more clean and readable.

Signed-off-by: Tushar Dave <redacted>
Tested-by: Aaron Brown <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---
 drivers/net/ethernet/intel/e1000e/82571.c |   14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/intel/e1000e/82571.c b/drivers/net/ethernet/intel/e1000e/82571.c
index 36db4df..19f4cb9 100644
--- a/drivers/net/ethernet/intel/e1000e/82571.c
+++ b/drivers/net/ethernet/intel/e1000e/82571.c
@@ -1677,16 +1677,18 @@ static s32 e1000_check_for_serdes_link_82571(struct e1000_hw *hw)
 			e_dbg("ANYSTATE  -> DOWN\n");
 		} else {
 			/*
-			 * Check several times, if Sync and Config
-			 * both are consistently 1 then simply ignore
-			 * the Invalid bit and restart Autoneg
+			 * Check several times, if SYNCH bit and CONFIG
+			 * bit both are consistently 1 then simply ignore
+			 * the IV bit and restart Autoneg
 			 */
 			for (i = 0; i < AN_RETRY_COUNT; i++) {
 				udelay(10);
 				rxcw = er32(RXCW);
-				if ((rxcw & E1000_RXCW_IV) &&
-				    !((rxcw & E1000_RXCW_SYNCH) &&
-				      (rxcw & E1000_RXCW_C))) {
+				if ((rxcw & E1000_RXCW_SYNCH) &&
+				    (rxcw & E1000_RXCW_C))
+					continue;
+
+				if (rxcw & E1000_RXCW_IV) {
 					mac->serdes_has_link = false;
 					mac->serdes_link_state =
 					    e1000_serdes_link_down;
-- 
1.7.10.4

[net-next 2/7] e1000e: Program the correct register for ITR when using MSI-X.

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:33

From: Matthew Vick <redacted>

When configuring interrupt throttling on 82574 in MSI-X mode, we need to
be programming the EITR registers instead of the ITR register.

-rc2: Renamed e1000_write_itr() to e1000e_write_itr(), fixed whitespace
      issues, and removed unnecessary !! operation.
-rc3: Reduced the scope of the loop variable in e1000e_write_itr().

Signed-off-by: Matthew Vick <redacted>
Acked-by: Bruce Allan <redacted>
Tested-by: Aaron Brown <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---
 drivers/net/ethernet/intel/e1000e/e1000.h   |    1 +
 drivers/net/ethernet/intel/e1000e/ethtool.c |    5 ++---
 drivers/net/ethernet/intel/e1000e/netdev.c  |   32 +++++++++++++++++++++++----
 3 files changed, 31 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h
index 6e6fffb..cd15332 100644
--- a/drivers/net/ethernet/intel/e1000e/e1000.h
+++ b/drivers/net/ethernet/intel/e1000e/e1000.h
@@ -514,6 +514,7 @@ extern void e1000e_set_interrupt_capability(struct e1000_adapter *adapter);
 extern void e1000e_reset_interrupt_capability(struct e1000_adapter *adapter);
 extern void e1000e_get_hw_control(struct e1000_adapter *adapter);
 extern void e1000e_release_hw_control(struct e1000_adapter *adapter);
+extern void e1000e_write_itr(struct e1000_adapter *adapter, u32 itr);
 
 extern unsigned int copybreak;
 
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index 905e214..105d554 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -1897,7 +1897,6 @@ static int e1000_set_coalesce(struct net_device *netdev,
 			      struct ethtool_coalesce *ec)
 {
 	struct e1000_adapter *adapter = netdev_priv(netdev);
-	struct e1000_hw *hw = &adapter->hw;
 
 	if ((ec->rx_coalesce_usecs > E1000_MAX_ITR_USECS) ||
 	    ((ec->rx_coalesce_usecs > 4) &&
@@ -1916,9 +1915,9 @@ static int e1000_set_coalesce(struct net_device *netdev,
 	}
 
 	if (adapter->itr_setting != 0)
-		ew32(ITR, 1000000000 / (adapter->itr * 256));
+		e1000e_write_itr(adapter, adapter->itr);
 	else
-		ew32(ITR, 0);
+		e1000e_write_itr(adapter, 0);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c
index ca477e8..95b2453 100644
--- a/drivers/net/ethernet/intel/e1000e/netdev.c
+++ b/drivers/net/ethernet/intel/e1000e/netdev.c
@@ -2474,6 +2474,30 @@ set_itr_now:
 }
 
 /**
+ * e1000e_write_itr - write the ITR value to the appropriate registers
+ * @adapter: address of board private structure
+ * @itr: new ITR value to program
+ *
+ * e1000e_write_itr determines if the adapter is in MSI-X mode
+ * and, if so, writes the EITR registers with the ITR value.
+ * Otherwise, it writes the ITR value into the ITR register.
+ **/
+void e1000e_write_itr(struct e1000_adapter *adapter, u32 itr)
+{
+	struct e1000_hw *hw = &adapter->hw;
+	u32 new_itr = itr ? 1000000000 / (itr * 256) : 0;
+
+	if (adapter->msix_entries) {
+		int vector;
+
+		for (vector = 0; vector < adapter->num_vectors; vector++)
+			writel(new_itr, hw->hw_addr + E1000_EITR_82574(vector));
+	} else {
+		ew32(ITR, new_itr);
+	}
+}
+
+/**
  * e1000_alloc_queues - Allocate memory for all rings
  * @adapter: board private structure to initialize
  **/
@@ -3059,7 +3083,7 @@ static void e1000_configure_rx(struct e1000_adapter *adapter)
 	/* irq moderation */
 	ew32(RADV, adapter->rx_abs_int_delay);
 	if ((adapter->itr_setting != 0) && (adapter->itr != 0))
-		ew32(ITR, 1000000000 / (adapter->itr * 256));
+		e1000e_write_itr(adapter, adapter->itr);
 
 	ctrl_ext = er32(CTRL_EXT);
 	/* Auto-Mask interrupts upon ICR access */
@@ -3486,14 +3510,14 @@ void e1000e_reset(struct e1000_adapter *adapter)
 				dev_info(&adapter->pdev->dev,
 					"Interrupt Throttle Rate turned off\n");
 				adapter->flags2 |= FLAG2_DISABLE_AIM;
-				ew32(ITR, 0);
+				e1000e_write_itr(adapter, 0);
 			}
 		} else if (adapter->flags2 & FLAG2_DISABLE_AIM) {
 			dev_info(&adapter->pdev->dev,
 				 "Interrupt Throttle Rate turned on\n");
 			adapter->flags2 &= ~FLAG2_DISABLE_AIM;
 			adapter->itr = 20000;
-			ew32(ITR, 1000000000 / (adapter->itr * 256));
+			e1000e_write_itr(adapter, adapter->itr);
 		}
 	}
 
@@ -4576,7 +4600,7 @@ link_up:
 			    adapter->gorc - adapter->gotc) / 10000;
 		u32 itr = goc > 0 ? (dif * 6000 / goc + 2000) : 8000;
 
-		ew32(ITR, 1000000000 / (itr * 256));
+		e1000e_write_itr(adapter, itr);
 	}
 
 	/* Cause software interrupt to ensure Rx ring is cleaned */
-- 
1.7.10.4

[net-next 3/7] ixgbe: Simplify logic for getting traffic class from user priority

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:41

From: Alexander Duyck <redacted>

This patch is meant to help simplify the logic for getting traffic classes
from user priorities. To do this I am adding a function named
ixgbe_dcb_get_tc_from_up that will go through the traffic classes in
reverse order in order to determine which traffic class contains a bit for
a given user priority.

Adding a declaration for this new function to the header so that
we have a centralized means for sorting out traffic classes belonging to
features such as FCoE.

Signed-off-by: Alexander Duyck <redacted>
Tested-by: Phil Schmitt <redacted>
Tested-by: Ross Brattain <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c |   30 ++++++++++++++++++++------
 1 file changed, 23 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
index 8bfaaee..4fd5a0d 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
@@ -231,16 +231,32 @@ void ixgbe_dcb_unpack_prio(struct ixgbe_dcb_config *cfg, int direction,
 	}
 }
 
-void ixgbe_dcb_unpack_map(struct ixgbe_dcb_config *cfg, int direction, u8 *map)
+static u8 ixgbe_dcb_get_tc_from_up(struct ixgbe_dcb_config *cfg,
+				   int direction, u8 up)
 {
-	int i, up;
-	unsigned long bitmap;
+	struct tc_configuration *tc_config = &cfg->tc_config[0];
+	u8 prio_mask = 1 << up;
+	u8 tc;
 
-	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
-		bitmap = cfg->tc_config[i].path[direction].up_to_tc_bitmap;
-		for_each_set_bit(up, &bitmap, MAX_USER_PRIORITY)
-			map[up] = i;
+	/*
+	 * Test for TCs 7 through 1 and report the first match we find.  If
+	 * we find no match we can assume that the TC is 0 since the TC must
+	 * be set for all user priorities
+	 */
+	for (tc = MAX_TRAFFIC_CLASS - 1; tc; tc--) {
+		if (prio_mask & tc_config[tc].path[direction].up_to_tc_bitmap)
+			break;
 	}
+
+	return tc;
+}
+
+void ixgbe_dcb_unpack_map(struct ixgbe_dcb_config *cfg, int direction, u8 *map)
+{
+	u8 up;
+
+	for (up = 0; up < MAX_USER_PRIORITY; up++)
+		map[up] = ixgbe_dcb_get_tc_from_up(cfg, direction, up);
 }
 
 /**
-- 
1.7.10.4

[net-next 4/7] ixgbe: Cleanup unpacking code for DCB

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:43

From: Alexander Duyck <redacted>

This is meant to be a generic clean-up of the remaining functions for
unpacking data from the DCB structures. The only real changes are:
replaced the variable i with tc for functions that were looping through the
traffic classes, and added a pointer for tc_class instead of path since
that way we only need to pull the pointer once instead of once per loop.

Signed-off-by: Alexander Duyck <redacted>
Tested-by: Phil Schmitt <redacted>
Tested-by: Ross Brattain <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c |   47 ++++++++++++--------------
 1 file changed, 22 insertions(+), 25 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
index 4fd5a0d..39ac2fe 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
@@ -180,55 +180,52 @@ out:
 
 void ixgbe_dcb_unpack_pfc(struct ixgbe_dcb_config *cfg, u8 *pfc_en)
 {
-	int i;
+	struct tc_configuration *tc_config = &cfg->tc_config[0];
+	int tc;
 
-	*pfc_en = 0;
-	for (i = 0; i < MAX_TRAFFIC_CLASS; i++)
-		*pfc_en |= !!(cfg->tc_config[i].dcb_pfc & 0xF) << i;
+	for (*pfc_en = 0, tc = 0; tc < MAX_TRAFFIC_CLASS; tc++) {
+		if (tc_config[tc].dcb_pfc != pfc_disabled)
+			*pfc_en |= 1 << tc;
+	}
 }
 
 void ixgbe_dcb_unpack_refill(struct ixgbe_dcb_config *cfg, int direction,
 			     u16 *refill)
 {
-	struct tc_bw_alloc *p;
-	int i;
+	struct tc_configuration *tc_config = &cfg->tc_config[0];
+	int tc;
 
-	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
-		p = &cfg->tc_config[i].path[direction];
-		refill[i] = p->data_credits_refill;
-	}
+	for (tc = 0; tc < MAX_TRAFFIC_CLASS; tc++)
+		refill[tc] = tc_config[tc].path[direction].data_credits_refill;
 }
 
 void ixgbe_dcb_unpack_max(struct ixgbe_dcb_config *cfg, u16 *max)
 {
-	int i;
+	struct tc_configuration *tc_config = &cfg->tc_config[0];
+	int tc;
 
-	for (i = 0; i < MAX_TRAFFIC_CLASS; i++)
-		max[i] = cfg->tc_config[i].desc_credits_max;
+	for (tc = 0; tc < MAX_TRAFFIC_CLASS; tc++)
+		max[tc] = tc_config[tc].desc_credits_max;
 }
 
 void ixgbe_dcb_unpack_bwgid(struct ixgbe_dcb_config *cfg, int direction,
 			    u8 *bwgid)
 {
-	struct tc_bw_alloc *p;
-	int i;
+	struct tc_configuration *tc_config = &cfg->tc_config[0];
+	int tc;
 
-	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
-		p = &cfg->tc_config[i].path[direction];
-		bwgid[i] = p->bwg_id;
-	}
+	for (tc = 0; tc < MAX_TRAFFIC_CLASS; tc++)
+		bwgid[tc] = tc_config[tc].path[direction].bwg_id;
 }
 
 void ixgbe_dcb_unpack_prio(struct ixgbe_dcb_config *cfg, int direction,
 			    u8 *ptype)
 {
-	struct tc_bw_alloc *p;
-	int i;
+	struct tc_configuration *tc_config = &cfg->tc_config[0];
+	int tc;
 
-	for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
-		p = &cfg->tc_config[i].path[direction];
-		ptype[i] = p->prio_type;
-	}
+	for (tc = 0; tc < MAX_TRAFFIC_CLASS; tc++)
+		ptype[tc] = tc_config[tc].path[direction].prio_type;
 }
 
 static u8 ixgbe_dcb_get_tc_from_up(struct ixgbe_dcb_config *cfg,
-- 
1.7.10.4

[net-next 5/7] ixgbe: Populate the prio_tc_map in ixgbe_setup_tc

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:45

From: Alexander Duyck <redacted>

There were cases where the prio_tc_map was not populated when we were
calling open.  This will result in us incorrectly configuring the traffic
classes when DCB is enabled.  In order to correct this I have updated the
code so that we now populate the values prior to allocating the q_vectors
and calling ixgbe_open.

Cc: John Fastabend <redacted>
Signed-off-by: Alexander Duyck <redacted>
Tested-by: Phil Schmitt <redacted>
Tested-by: Ross Brattain <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c    |    3 +--
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.h    |    1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c |   26 ++++-----------------
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c   |   28 +++++++++++++++++++++++
 4 files changed, 35 insertions(+), 23 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
index 39ac2fe..5442b35 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c
@@ -228,8 +228,7 @@ void ixgbe_dcb_unpack_prio(struct ixgbe_dcb_config *cfg, int direction,
 		ptype[tc] = tc_config[tc].path[direction].prio_type;
 }
 
-static u8 ixgbe_dcb_get_tc_from_up(struct ixgbe_dcb_config *cfg,
-				   int direction, u8 up)
+u8 ixgbe_dcb_get_tc_from_up(struct ixgbe_dcb_config *cfg, int direction, u8 up)
 {
 	struct tc_configuration *tc_config = &cfg->tc_config[0];
 	u8 prio_mask = 1 << up;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.h
index 24333b7..1f4108e 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.h
@@ -146,6 +146,7 @@ void ixgbe_dcb_unpack_max(struct ixgbe_dcb_config *, u16 *);
 void ixgbe_dcb_unpack_bwgid(struct ixgbe_dcb_config *, int, u8 *);
 void ixgbe_dcb_unpack_prio(struct ixgbe_dcb_config *, int, u8 *);
 void ixgbe_dcb_unpack_map(struct ixgbe_dcb_config *, int, u8 *);
+u8 ixgbe_dcb_get_tc_from_up(struct ixgbe_dcb_config *, int, u8);
 
 /* DCB credits calculation */
 s32 ixgbe_dcb_calculate_tc_credits(struct ixgbe_hw *,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c
index 5164a21..f1e002d 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_dcb_nl.c
@@ -151,34 +151,21 @@ static u8 ixgbe_dcbnl_get_state(struct net_device *netdev)
 
 static u8 ixgbe_dcbnl_set_state(struct net_device *netdev, u8 state)
 {
-	int err = 0;
-	u8 prio_tc[MAX_USER_PRIORITY] = {0};
-	int i;
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
+	int err = 0;
 
 	/* Fail command if not in CEE mode */
 	if (!(adapter->dcbx_cap & DCB_CAP_DCBX_VER_CEE))
 		return 1;
 
 	/* verify there is something to do, if not then exit */
-	if (!!state != !(adapter->flags & IXGBE_FLAG_DCB_ENABLED))
-		goto out;
-
-	if (state > 0) {
-		err = ixgbe_setup_tc(netdev, adapter->dcb_cfg.num_tcs.pg_tcs);
-		ixgbe_dcb_unpack_map(&adapter->dcb_cfg, DCB_TX_CONFIG, prio_tc);
-	} else {
-		err = ixgbe_setup_tc(netdev, 0);
-	}
-
-	if (err)
+	if (!state == !(adapter->flags & IXGBE_FLAG_DCB_ENABLED))
 		goto out;
 
-	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++)
-		netdev_set_prio_tc_map(netdev, i, prio_tc[i]);
-
+	err = ixgbe_setup_tc(netdev,
+			     state ? adapter->dcb_cfg.num_tcs.pg_tcs : 0);
 out:
-	return err ? 1 : 0;
+	return !!err;
 }
 
 static void ixgbe_dcbnl_get_perm_hw_addr(struct net_device *netdev,
@@ -584,9 +571,6 @@ static int ixgbe_dcbnl_ieee_setets(struct net_device *dev,
 	if (err)
 		goto err_out;
 
-	for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++)
-		netdev_set_prio_tc_map(dev, i, ets->prio_tc[i]);
-
 	err = ixgbe_dcb_hw_ets(&adapter->hw, ets, max_frame);
 err_out:
 	return err;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index d3cf887..91bc60f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -6596,6 +6596,31 @@ static void ixgbe_validate_rtr(struct ixgbe_adapter *adapter, u8 tc)
 }
 
 /**
+ * ixgbe_set_prio_tc_map - Configure netdev prio tc map
+ * @adapter: Pointer to adapter struct
+ *
+ * Populate the netdev user priority to tc map
+ */
+static void ixgbe_set_prio_tc_map(struct ixgbe_adapter *adapter)
+{
+	struct net_device *dev = adapter->netdev;
+	struct ixgbe_dcb_config *dcb_cfg = &adapter->dcb_cfg;
+	struct ieee_ets *ets = adapter->ixgbe_ieee_ets;
+	u8 prio;
+
+	for (prio = 0; prio < MAX_USER_PRIORITY; prio++) {
+		u8 tc = 0;
+
+		if (adapter->dcbx_cap & DCB_CAP_DCBX_VER_CEE)
+			tc = ixgbe_dcb_get_tc_from_up(dcb_cfg, 0, prio);
+		else if (ets)
+			tc = ets->prio_tc[prio];
+
+		netdev_set_prio_tc_map(dev, prio, tc);
+	}
+}
+
+/**
  * ixgbe_setup_tc - configure net_device for multiple traffic classes
  *
  * @netdev: net device to configure
@@ -6633,6 +6658,8 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 
 	if (tc) {
 		netdev_set_num_tc(dev, tc);
+		ixgbe_set_prio_tc_map(adapter);
+
 		adapter->flags |= IXGBE_FLAG_DCB_ENABLED;
 		adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
 
@@ -6642,6 +6669,7 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 		}
 	} else {
 		netdev_reset_tc(dev);
+
 		if (adapter->hw.mac.type == ixgbe_mac_82598EB)
 			adapter->hw.fc.requested_mode = adapter->last_lfc_mode;
 
-- 
1.7.10.4

[net-next 6/7] ixgbe: Add function for obtaining FCoE TC based on FCoE user priority

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:48

From: Alexander Duyck <redacted>

In upcoming patches it will become increasingly common to need to determine
the FCoE traffic class in order to determine the correct queues for FCoE.
In order to make this easier I am adding a function for obtaining the FCoE
traffic class based on the user priority.

Cc: John Fastabend <redacted>
Signed-off-by: Alexander Duyck <redacted>
Tested-by: Phil Schmitt <redacted>
Tested-by: Ross Brattain <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe.h      |    1 +
 drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c |   15 +++++++++++++++
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c  |    7 ++-----
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |   14 ++++----------
 4 files changed, 22 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 2ffdc8f..5a75a9c 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -707,6 +707,7 @@ extern u8 ixgbe_fcoe_setapp(struct ixgbe_adapter *adapter, u8 up);
 extern int ixgbe_fcoe_get_wwn(struct net_device *netdev, u64 *wwn, int type);
 extern int ixgbe_fcoe_get_hbainfo(struct net_device *netdev,
 				  struct netdev_fcoe_hbainfo *info);
+extern u8 ixgbe_fcoe_get_tc(struct ixgbe_adapter *adapter);
 #endif /* IXGBE_FCOE */
 
 static inline struct netdev_queue *txring_txq(const struct ixgbe_ring *ring)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
index 0922ece..cc28c44 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
@@ -960,3 +960,18 @@ int ixgbe_fcoe_get_hbainfo(struct net_device *netdev,
 
 	return 0;
 }
+
+/**
+ * ixgbe_fcoe_get_tc - get the current TC that fcoe is mapped to
+ * @adapter - pointer to the device adapter structure
+ *
+ * Return : TC that FCoE is mapped to
+ */
+u8 ixgbe_fcoe_get_tc(struct ixgbe_adapter *adapter)
+{
+#ifdef CONFIG_IXGBE_DCB
+	return netdev_get_prio_tc_map(adapter->netdev, adapter->fcoe.up);
+#else
+	return 0;
+#endif
+}
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
index 83eadd0..c0ff31e 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
@@ -49,8 +49,8 @@ static inline bool ixgbe_cache_ring_rss(struct ixgbe_adapter *adapter)
 
 	return true;
 }
-#ifdef CONFIG_IXGBE_DCB
 
+#ifdef CONFIG_IXGBE_DCB
 /* ixgbe_get_first_reg_idx - Return first register index associated with ring */
 static void ixgbe_get_first_reg_idx(struct ixgbe_adapter *adapter, u8 tc,
 				    unsigned int *tx, unsigned int *rx)
@@ -343,13 +343,10 @@ static inline bool ixgbe_set_dcb_queues(struct ixgbe_adapter *adapter)
 	 * configuration later.
 	 */
 	if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) {
-		u8 prio_tc[MAX_USER_PRIORITY] = {0};
-		int tc;
+		u8 tc = ixgbe_fcoe_get_tc(adapter);
 		struct ixgbe_ring_feature *f =
 					&adapter->ring_feature[RING_F_FCOE];
 
-		ixgbe_dcb_unpack_map(&adapter->dcb_cfg, DCB_TX_CONFIG, prio_tc);
-		tc = prio_tc[adapter->fcoe.up];
 		f->indices = dev->tc_to_txq[tc].count;
 		f->offset = dev->tc_to_txq[tc].offset;
 	}
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 91bc60f..3d7ce7e 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -3646,18 +3646,12 @@ static int ixgbe_hpbthresh(struct ixgbe_adapter *adapter, int pb)
 
 #ifdef IXGBE_FCOE
 	/* FCoE traffic class uses FCOE jumbo frames */
-	if (dev->features & NETIF_F_FCOE_MTU) {
-		int fcoe_pb = 0;
-
-#ifdef CONFIG_IXGBE_DCB
-		fcoe_pb = netdev_get_prio_tc_map(dev, adapter->fcoe.up);
+	if ((dev->features & NETIF_F_FCOE_MTU) &&
+	    (tc < IXGBE_FCOE_JUMBO_FRAME_SIZE) &&
+	    (pb == ixgbe_fcoe_get_tc(adapter)))
+		tc = IXGBE_FCOE_JUMBO_FRAME_SIZE;
 
 #endif
-		if (fcoe_pb == pb && tc < IXGBE_FCOE_JUMBO_FRAME_SIZE)
-			tc = IXGBE_FCOE_JUMBO_FRAME_SIZE;
-	}
-#endif
-
 	/* Calculate delay value for device */
 	switch (hw->mac.type) {
 	case ixgbe_mac_X540:
-- 
1.7.10.4

[net-next 7/7] ixgbe: Merge FCoE set_num and cache_ring calls into RSS/DCB config

From: Jeff Kirsher <hidden>
Date: 2012-07-14 23:07:52

From: Alexander Duyck <redacted>

This change merges the ixgbe_cache_ring_fcoe and ixgbe_set_fcoe_queues
logic into the DCB and RSS initialization calls.

Cc: John Fastabend <redacted>
Signed-off-by: Alexander Duyck <redacted>
Tested-by: Phil Schmitt <redacted>
Tested-by: Ross Brattain <redacted>
Signed-off-by: Jeff Kirsher <redacted>
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c  |  257 +++++++++++--------------
 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c |   19 +-
 2 files changed, 129 insertions(+), 147 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
index c0ff31e..d308e71 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c
@@ -28,28 +28,6 @@
 #include "ixgbe.h"
 #include "ixgbe_sriov.h"
 
-/**
- * ixgbe_cache_ring_rss - Descriptor ring to register mapping for RSS
- * @adapter: board private structure to initialize
- *
- * Cache the descriptor ring offsets for RSS to the assigned rings.
- *
- **/
-static inline bool ixgbe_cache_ring_rss(struct ixgbe_adapter *adapter)
-{
-	int i;
-
-	if (!(adapter->flags & IXGBE_FLAG_RSS_ENABLED))
-		return false;
-
-	for (i = 0; i < adapter->num_rx_queues; i++)
-		adapter->rx_ring[i]->reg_idx = i;
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		adapter->tx_ring[i]->reg_idx = i;
-
-	return true;
-}
-
 #ifdef CONFIG_IXGBE_DCB
 /* ixgbe_get_first_reg_idx - Return first register index associated with ring */
 static void ixgbe_get_first_reg_idx(struct ixgbe_adapter *adapter, u8 tc,
@@ -136,39 +114,8 @@ static inline bool ixgbe_cache_ring_dcb(struct ixgbe_adapter *adapter)
 
 	return true;
 }
-#endif
-
-#ifdef IXGBE_FCOE
-/**
- * ixgbe_cache_ring_fcoe - Descriptor ring to register mapping for the FCoE
- * @adapter: board private structure to initialize
- *
- * Cache the descriptor ring offsets for FCoE mode to the assigned rings.
- *
- */
-static inline bool ixgbe_cache_ring_fcoe(struct ixgbe_adapter *adapter)
-{
-	struct ixgbe_ring_feature *f = &adapter->ring_feature[RING_F_FCOE];
-	int i;
-	u8 fcoe_rx_i = 0, fcoe_tx_i = 0;
-
-	if (!(adapter->flags & IXGBE_FLAG_FCOE_ENABLED))
-		return false;
-
-	if (adapter->flags & IXGBE_FLAG_RSS_ENABLED) {
-		ixgbe_cache_ring_rss(adapter);
-
-		fcoe_rx_i = f->offset;
-		fcoe_tx_i = f->offset;
-	}
-	for (i = 0; i < f->indices; i++, fcoe_rx_i++, fcoe_tx_i++) {
-		adapter->rx_ring[f->offset + i]->reg_idx = fcoe_rx_i;
-		adapter->tx_ring[f->offset + i]->reg_idx = fcoe_tx_i;
-	}
-	return true;
-}
 
-#endif /* IXGBE_FCOE */
+#endif
 /**
  * ixgbe_cache_ring_sriov - Descriptor ring to register mapping for sriov
  * @adapter: board private structure to initialize
@@ -188,6 +135,28 @@ static inline bool ixgbe_cache_ring_sriov(struct ixgbe_adapter *adapter)
 }
 
 /**
+ * ixgbe_cache_ring_rss - Descriptor ring to register mapping for RSS
+ * @adapter: board private structure to initialize
+ *
+ * Cache the descriptor ring offsets for RSS to the assigned rings.
+ *
+ **/
+static bool ixgbe_cache_ring_rss(struct ixgbe_adapter *adapter)
+{
+	int i;
+
+	if (!(adapter->flags & IXGBE_FLAG_RSS_ENABLED))
+		return false;
+
+	for (i = 0; i < adapter->num_rx_queues; i++)
+		adapter->rx_ring[i]->reg_idx = i;
+	for (i = 0; i < adapter->num_tx_queues; i++)
+		adapter->tx_ring[i]->reg_idx = i;
+
+	return true;
+}
+
+/**
  * ixgbe_cache_ring_register - Descriptor ring to register mapping
  * @adapter: board private structure to initialize
  *
@@ -212,13 +181,7 @@ static void ixgbe_cache_ring_register(struct ixgbe_adapter *adapter)
 		return;
 #endif
 
-#ifdef IXGBE_FCOE
-	if (ixgbe_cache_ring_fcoe(adapter))
-		return;
-#endif /* IXGBE_FCOE */
-
-	if (ixgbe_cache_ring_rss(adapter))
-		return;
+	ixgbe_cache_ring_rss(adapter);
 }
 
 /**
@@ -234,6 +197,74 @@ static inline bool ixgbe_set_sriov_queues(struct ixgbe_adapter *adapter)
 	return false;
 }
 
+#define IXGBE_RSS_16Q_MASK	0xF
+#define IXGBE_RSS_8Q_MASK	0x7
+#define IXGBE_RSS_4Q_MASK	0x3
+#define IXGBE_RSS_2Q_MASK	0x1
+#define IXGBE_RSS_DISABLED_MASK	0x0
+
+#ifdef CONFIG_IXGBE_DCB
+static bool ixgbe_set_dcb_queues(struct ixgbe_adapter *adapter)
+{
+	struct net_device *dev = adapter->netdev;
+	struct ixgbe_ring_feature *f;
+	int rss_i, rss_m, i;
+	int tcs;
+
+	/* Map queue offset and counts onto allocated tx queues */
+	tcs = netdev_get_num_tc(dev);
+
+	/* verify we have DCB queueing enabled before proceeding */
+	if (tcs <= 1)
+		return false;
+
+	/* determine the upper limit for our current DCB mode */
+	rss_i = dev->num_tx_queues / tcs;
+	if (adapter->hw.mac.type == ixgbe_mac_82598EB) {
+		/* 8 TC w/ 4 queues per TC */
+		rss_i = min_t(u16, rss_i, 4);
+		rss_m = IXGBE_RSS_4Q_MASK;
+	} else if (tcs > 4) {
+		/* 8 TC w/ 8 queues per TC */
+		rss_i = min_t(u16, rss_i, 8);
+		rss_m = IXGBE_RSS_8Q_MASK;
+	} else {
+		/* 4 TC w/ 16 queues per TC */
+		rss_i = min_t(u16, rss_i, 16);
+		rss_m = IXGBE_RSS_16Q_MASK;
+	}
+
+	/* set RSS mask and indices */
+	f = &adapter->ring_feature[RING_F_RSS];
+	rss_i = min_t(int, rss_i, f->limit);
+	f->indices = rss_i;
+	f->mask = rss_m;
+
+#ifdef IXGBE_FCOE
+	/* FCoE enabled queues require special configuration indexed
+	 * by feature specific indices and offset. Here we map FCoE
+	 * indices onto the DCB queue pairs allowing FCoE to own
+	 * configuration later.
+	 */
+	if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) {
+		u8 tc = ixgbe_fcoe_get_tc(adapter);
+
+		f = &adapter->ring_feature[RING_F_FCOE];
+		f->indices = min_t(u16, rss_i, f->limit);
+		f->offset = rss_i * tc;
+	}
+
+#endif /* IXGBE_FCOE */
+	for (i = 0; i < tcs; i++)
+		netdev_set_tc_queue(dev, i, rss_i, rss_i * i);
+
+	adapter->num_tx_queues = rss_i * tcs;
+	adapter->num_rx_queues = rss_i * tcs;
+
+	return true;
+}
+
+#endif
 /**
  * ixgbe_set_rss_queues - Allocate queues for RSS
  * @adapter: board private structure to initialize
@@ -257,7 +288,7 @@ static bool ixgbe_set_rss_queues(struct ixgbe_adapter *adapter)
 	rss_i = f->limit;
 
 	f->indices = rss_i;
-	f->mask = 0xF;
+	f->mask = IXGBE_RSS_16Q_MASK;
 
 	/*
 	 * Use Flow Director in addition to RSS to ensure the best
@@ -271,90 +302,41 @@ static bool ixgbe_set_rss_queues(struct ixgbe_adapter *adapter)
 		rss_i = max_t(u16, rss_i, f->indices);
 	}
 
-	adapter->num_rx_queues = rss_i;
-	adapter->num_tx_queues = rss_i;
-
-	return true;
-}
-
 #ifdef IXGBE_FCOE
-/**
- * ixgbe_set_fcoe_queues - Allocate queues for Fiber Channel over Ethernet (FCoE)
- * @adapter: board private structure to initialize
- *
- * FCoE RX FCRETA can use up to 8 rx queues for up to 8 different exchanges.
- * Offset is used as the index of the first rx queue used by FCoE.
- **/
-static inline bool ixgbe_set_fcoe_queues(struct ixgbe_adapter *adapter)
-{
-	struct ixgbe_ring_feature *f = &adapter->ring_feature[RING_F_FCOE];
+	/*
+	 * FCoE can exist on the same rings as standard network traffic
+	 * however it is preferred to avoid that if possible.  In order
+	 * to get the best performance we allocate as many FCoE queues
+	 * as we can and we place them at the end of the ring array to
+	 * avoid sharing queues with standard RSS on systems with 24 or
+	 * more CPUs.
+	 */
+	if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) {
+		struct net_device *dev = adapter->netdev;
+		u16 fcoe_i;
 
-	if (!(adapter->flags & IXGBE_FLAG_FCOE_ENABLED))
-		return false;
+		f = &adapter->ring_feature[RING_F_FCOE];
 
-	f->indices = min_t(int, num_online_cpus(), f->limit);
+		/* merge FCoE queues with RSS queues */
+		fcoe_i = min_t(u16, f->limit + rss_i, num_online_cpus());
+		fcoe_i = min_t(u16, fcoe_i, dev->num_tx_queues);
 
-	adapter->num_rx_queues = 1;
-	adapter->num_tx_queues = 1;
+		/* limit indices to rss_i if MSI-X is disabled */
+		if (!(adapter->flags & IXGBE_FLAG_MSIX_ENABLED))
+			fcoe_i = rss_i;
 
-	if (adapter->flags & IXGBE_FLAG_RSS_ENABLED) {
-		e_info(probe, "FCoE enabled with RSS\n");
-		ixgbe_set_rss_queues(adapter);
+		/* attempt to reserve some queues for just FCoE */
+		f->indices = min_t(u16, fcoe_i, f->limit);
+		f->offset = fcoe_i - f->indices;
+		rss_i = max_t(u16, fcoe_i, rss_i);
 	}
 
-	/* adding FCoE rx rings to the end */
-	f->offset = adapter->num_rx_queues;
-	adapter->num_rx_queues += f->indices;
-	adapter->num_tx_queues += f->indices;
-
-	return true;
-}
 #endif /* IXGBE_FCOE */
-
-/* Artificial max queue cap per traffic class in DCB mode */
-#define DCB_QUEUE_CAP 8
-
-#ifdef CONFIG_IXGBE_DCB
-static inline bool ixgbe_set_dcb_queues(struct ixgbe_adapter *adapter)
-{
-	int per_tc_q, q, i, offset = 0;
-	struct net_device *dev = adapter->netdev;
-	int tcs = netdev_get_num_tc(dev);
-
-	if (!tcs)
-		return false;
-
-	/* Map queue offset and counts onto allocated tx queues */
-	per_tc_q = min_t(unsigned int, dev->num_tx_queues / tcs, DCB_QUEUE_CAP);
-	q = min_t(int, num_online_cpus(), per_tc_q);
-
-	for (i = 0; i < tcs; i++) {
-		netdev_set_tc_queue(dev, i, q, offset);
-		offset += q;
-	}
-
-	adapter->num_tx_queues = q * tcs;
-	adapter->num_rx_queues = q * tcs;
-
-#ifdef IXGBE_FCOE
-	/* FCoE enabled queues require special configuration indexed
-	 * by feature specific indices and offset. Here we map FCoE
-	 * indices onto the DCB queue pairs allowing FCoE to own
-	 * configuration later.
-	 */
-	if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) {
-		u8 tc = ixgbe_fcoe_get_tc(adapter);
-		struct ixgbe_ring_feature *f =
-					&adapter->ring_feature[RING_F_FCOE];
-
-		f->indices = dev->tc_to_txq[tc].count;
-		f->offset = dev->tc_to_txq[tc].offset;
-	}
-#endif
+	adapter->num_rx_queues = rss_i;
+	adapter->num_tx_queues = rss_i;
 
 	return true;
 }
-#endif
 
 /**
  * ixgbe_set_num_queues - Allocate queues for device, feature dependent
@@ -383,11 +365,6 @@ static int ixgbe_set_num_queues(struct ixgbe_adapter *adapter)
 		goto done;
 
 #endif
-#ifdef IXGBE_FCOE
-	if (ixgbe_set_fcoe_queues(adapter))
-		goto done;
-
-#endif /* IXGBE_FCOE */
 	if (ixgbe_set_rss_queues(adapter))
 		goto done;
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 3d7ce7e..ee230f5 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -3610,16 +3610,17 @@ static void ixgbe_configure_dcb(struct ixgbe_adapter *adapter)
 	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;
 
-		for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
-			u8 msb = 0;
-			u8 cnt = adapter->netdev->tc_to_txq[i].count;
-
-			while (cnt >>= 1)
-				msb++;
+		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);
 	}
 }
@@ -7027,7 +7028,11 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 #endif
 
 	if (ii->mac == ixgbe_mac_82598EB)
+#ifdef CONFIG_IXGBE_DCB
+		indices = min_t(unsigned int, indices, MAX_TRAFFIC_CLASS * 4);
+#else
 		indices = min_t(unsigned int, indices, IXGBE_MAX_RSS_INDICES);
+#endif
 	else
 		indices = min_t(unsigned int, indices, IXGBE_MAX_FDIR_INDICES);
 
-- 
1.7.10.4

Re: [net-next 0/7][pull request] Intel Wired LAN Driver Updates

From: David Miller <davem@davemloft.net>
Date: 2012-07-17 06:04:44

From: Jeff Kirsher <redacted>
Date: Sat, 14 Jul 2012 16:06:58 -0700
This series contains updates to e1000e and ixgbe.

The following are changes since commit 141e369de698f2e17bf716b83fcc647ddcb2220c:
  xfrm: Initialize the struct xfrm_dst behind the dst_enty field
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next master

Alexander Duyck (5):
  ixgbe: Simplify logic for getting traffic class from user priority
  ixgbe: Cleanup unpacking code for DCB
  ixgbe: Populate the prio_tc_map in ixgbe_setup_tc
  ixgbe: Add function for obtaining FCoE TC based on FCoE user priority
  ixgbe: Merge FCoE set_num and cache_ring calls into RSS/DCB config

Matthew Vick (1):
  e1000e: Program the correct register for ITR when using MSI-X.

Tushar Dave (1):
  e1000e: Cleanup code logic in e1000_check_for_serdes_link_82571()
Pulled, thanks Jeff.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help