* [net-next-2.6 13/16] ixgbe: fix namespace issue with ixgbe_dcb_txq_to_tc
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: Don Skidmore, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Don Skidmore <donald.c.skidmore@intel.com>
We didn't need the prototype and it was causing namespace complaints so
I made it static.
Signed-off-by: Don Skidmore <donald.c.skidmore@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.h | 1 -
drivers/net/ixgbe/ixgbe_main.c | 2 +-
2 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe.h b/drivers/net/ixgbe/ixgbe.h
index d04afde..12769b5 100644
--- a/drivers/net/ixgbe/ixgbe.h
+++ b/drivers/net/ixgbe/ixgbe.h
@@ -525,7 +525,6 @@ extern void ixgbe_unmap_and_free_tx_resource(struct ixgbe_ring *,
extern void ixgbe_alloc_rx_buffers(struct ixgbe_ring *, u16);
extern void ixgbe_write_eitr(struct ixgbe_q_vector *);
extern int ethtool_ioctl(struct ifreq *ifr);
-extern u8 ixgbe_dcb_txq_to_tc(struct ixgbe_adapter *adapter, u8 index);
extern s32 ixgbe_reinit_fdir_tables_82599(struct ixgbe_hw *hw);
extern s32 ixgbe_init_fdir_signature_82599(struct ixgbe_hw *hw, u32 pballoc);
extern s32 ixgbe_init_fdir_perfect_82599(struct ixgbe_hw *hw, u32 pballoc);
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index ef4a517..f5bdc0a 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -648,7 +648,7 @@ void ixgbe_unmap_and_free_tx_resource(struct ixgbe_ring *tx_ring,
*
* Returns : a tc index for use in range 0-7, or 0-3
*/
-u8 ixgbe_dcb_txq_to_tc(struct ixgbe_adapter *adapter, u8 reg_idx)
+static u8 ixgbe_dcb_txq_to_tc(struct ixgbe_adapter *adapter, u8 reg_idx)
{
int tc = -1;
int dcb_i = adapter->ring_feature[RING_F_DCB].indices;
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 11/16] ixgbe: DCB, remove RESET bit it is no longer needed
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: John Fastabend, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: John Fastabend <john.r.fastabend@intel.com>
This removes the RESET bit previously used to force a device
reset when DCB bandwidth configurations were changed. This can
now be done dynamically without a reset so the bit is no longer
needed. The only remaining operations that force a device reset
are DCB enable/disable and FCoE application priority changes.
DCB enable/disable is a hardware requirement.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_dcb_nl.c | 37 ++++++++++++-------------------------
1 files changed, 12 insertions(+), 25 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ixgbe/ixgbe_dcb_nl.c
index b3a8d24..c94adec 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_nl.c
@@ -37,7 +37,6 @@
#define BIT_PG_RX 0x04
#define BIT_PG_TX 0x08
#define BIT_APP_UPCHG 0x10
-#define BIT_RESETLINK 0x40
#define BIT_LINKSPEED 0x80
/* Responses for the DCB_C_SET_ALL command */
@@ -345,7 +344,6 @@ static void ixgbe_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority,
static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
{
struct ixgbe_adapter *adapter = netdev_priv(netdev);
- bool do_reset;
int ret;
if (!adapter->dcb_set_bitmap)
@@ -358,23 +356,17 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
return DCB_NO_HW_CHG;
/*
- * Only take down the adapter if the configuration change
- * requires a reset.
+ * Only take down the adapter if an app change occured. FCoE
+ * may shuffle tx rings in this case and this can not be done
+ * without a reset currently.
*/
- do_reset = adapter->dcb_set_bitmap & (BIT_RESETLINK | BIT_APP_UPCHG);
-
- if (do_reset) {
+ if (adapter->dcb_set_bitmap & BIT_APP_UPCHG) {
while (test_and_set_bit(__IXGBE_RESETTING, &adapter->state))
msleep(1);
- if (adapter->dcb_set_bitmap & BIT_APP_UPCHG) {
- if (netif_running(netdev))
- netdev->netdev_ops->ndo_stop(netdev);
- ixgbe_clear_interrupt_scheme(adapter);
- } else {
- if (netif_running(netdev))
- ixgbe_down(adapter);
- }
+ if (netif_running(netdev))
+ netdev->netdev_ops->ndo_stop(netdev);
+ ixgbe_clear_interrupt_scheme(adapter);
}
if (adapter->dcb_cfg.pfc_mode_enable) {
@@ -403,15 +395,10 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
}
}
- if (do_reset) {
- if (adapter->dcb_set_bitmap & BIT_APP_UPCHG) {
- ixgbe_init_interrupt_scheme(adapter);
- if (netif_running(netdev))
- netdev->netdev_ops->ndo_open(netdev);
- } else {
- if (netif_running(netdev))
- ixgbe_up(adapter);
- }
+ if (adapter->dcb_set_bitmap & BIT_APP_UPCHG) {
+ ixgbe_init_interrupt_scheme(adapter);
+ if (netif_running(netdev))
+ netdev->netdev_ops->ndo_open(netdev);
ret = DCB_HW_CHG_RST;
}
@@ -456,7 +443,7 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
if (adapter->dcb_cfg.pfc_mode_enable)
adapter->hw.fc.current_mode = ixgbe_fc_pfc;
- if (do_reset)
+ if (adapter->dcb_set_bitmap & BIT_APP_UPCHG)
clear_bit(__IXGBE_RESETTING, &adapter->state);
adapter->dcb_set_bitmap = 0x00;
return ret;
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 10/16] ixgbe: DCB, do not reset on CEE pg changes
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: John Fastabend, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: John Fastabend <john.r.fastabend@intel.com>
The 82599 and 82598 devices do not require hardware resets to
configure CEE pg settings. This patch changes DCB configuration
to set the CEE pg values directly from the dcbnl ops routine.
This reduces the number of resets seen on the wire and allows
LLDP to reach a steady state faster.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_dcb.h | 4 +++
drivers/net/ixgbe/ixgbe_dcb_nl.c | 48 +++++++++++++++++++++++++++----------
2 files changed, 39 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_dcb.h b/drivers/net/ixgbe/ixgbe_dcb.h
index 4e4a641..e593511 100644
--- a/drivers/net/ixgbe/ixgbe_dcb.h
+++ b/drivers/net/ixgbe/ixgbe_dcb.h
@@ -148,6 +148,10 @@ struct ixgbe_dcb_config {
/* DCB driver APIs */
void ixgbe_dcb_unpack_pfc(struct ixgbe_dcb_config *cfg, u8 *pfc_en);
+void ixgbe_dcb_unpack_refill(struct ixgbe_dcb_config *, int, u16 *);
+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 *);
/* DCB credits calculation */
s32 ixgbe_ieee_credits(__u8 *bw, __u16 *refill, __u16 *max, int max_frame);
diff --git a/drivers/net/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ixgbe/ixgbe_dcb_nl.c
index e75a3c9..b3a8d24 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_nl.c
@@ -225,10 +225,8 @@ static void ixgbe_dcbnl_set_pg_tc_cfg_tx(struct net_device *netdev, int tc,
(adapter->temp_dcb_cfg.tc_config[tc].path[0].bwg_percent !=
adapter->dcb_cfg.tc_config[tc].path[0].bwg_percent) ||
(adapter->temp_dcb_cfg.tc_config[tc].path[0].up_to_tc_bitmap !=
- adapter->dcb_cfg.tc_config[tc].path[0].up_to_tc_bitmap)) {
+ adapter->dcb_cfg.tc_config[tc].path[0].up_to_tc_bitmap))
adapter->dcb_set_bitmap |= BIT_PG_TX;
- adapter->dcb_set_bitmap |= BIT_RESETLINK;
- }
}
static void ixgbe_dcbnl_set_pg_bwg_cfg_tx(struct net_device *netdev, int bwg_id,
@@ -239,10 +237,8 @@ static void ixgbe_dcbnl_set_pg_bwg_cfg_tx(struct net_device *netdev, int bwg_id,
adapter->temp_dcb_cfg.bw_percentage[0][bwg_id] = bw_pct;
if (adapter->temp_dcb_cfg.bw_percentage[0][bwg_id] !=
- adapter->dcb_cfg.bw_percentage[0][bwg_id]) {
+ adapter->dcb_cfg.bw_percentage[0][bwg_id])
adapter->dcb_set_bitmap |= BIT_PG_TX;
- adapter->dcb_set_bitmap |= BIT_RESETLINK;
- }
}
static void ixgbe_dcbnl_set_pg_tc_cfg_rx(struct net_device *netdev, int tc,
@@ -269,10 +265,8 @@ static void ixgbe_dcbnl_set_pg_tc_cfg_rx(struct net_device *netdev, int tc,
(adapter->temp_dcb_cfg.tc_config[tc].path[1].bwg_percent !=
adapter->dcb_cfg.tc_config[tc].path[1].bwg_percent) ||
(adapter->temp_dcb_cfg.tc_config[tc].path[1].up_to_tc_bitmap !=
- adapter->dcb_cfg.tc_config[tc].path[1].up_to_tc_bitmap)) {
+ adapter->dcb_cfg.tc_config[tc].path[1].up_to_tc_bitmap))
adapter->dcb_set_bitmap |= BIT_PG_RX;
- adapter->dcb_set_bitmap |= BIT_RESETLINK;
- }
}
static void ixgbe_dcbnl_set_pg_bwg_cfg_rx(struct net_device *netdev, int bwg_id,
@@ -283,10 +277,8 @@ static void ixgbe_dcbnl_set_pg_bwg_cfg_rx(struct net_device *netdev, int bwg_id,
adapter->temp_dcb_cfg.bw_percentage[1][bwg_id] = bw_pct;
if (adapter->temp_dcb_cfg.bw_percentage[1][bwg_id] !=
- adapter->dcb_cfg.bw_percentage[1][bwg_id]) {
+ adapter->dcb_cfg.bw_percentage[1][bwg_id])
adapter->dcb_set_bitmap |= BIT_PG_RX;
- adapter->dcb_set_bitmap |= BIT_RESETLINK;
- }
}
static void ixgbe_dcbnl_get_pg_tc_cfg_tx(struct net_device *netdev, int tc,
@@ -421,7 +413,9 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
ixgbe_up(adapter);
}
ret = DCB_HW_CHG_RST;
- } else if (adapter->dcb_set_bitmap & BIT_PFC) {
+ }
+
+ if (adapter->dcb_set_bitmap & BIT_PFC) {
u8 pfc_en;
ixgbe_dcb_unpack_pfc(&adapter->dcb_cfg, &pfc_en);
@@ -431,6 +425,34 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
ixgbe_dcb_config_pfc_82599(&adapter->hw, pfc_en);
ret = DCB_HW_CHG;
}
+
+ if (adapter->dcb_set_bitmap & (BIT_PG_TX|BIT_PG_RX)) {
+ u16 refill[MAX_TRAFFIC_CLASS], max[MAX_TRAFFIC_CLASS];
+ u8 bwg_id[MAX_TRAFFIC_CLASS], prio_type[MAX_TRAFFIC_CLASS];
+ int max_frame = adapter->netdev->mtu + ETH_HLEN + ETH_FCS_LEN;
+
+#ifdef CONFIG_FCOE
+ if (adapter->netdev->features & NETIF_F_FCOE_MTU)
+ max_frame = max(max_frame, IXGBE_FCOE_JUMBO_FRAME_SIZE);
+#endif
+
+ ixgbe_dcb_calculate_tc_credits(&adapter->hw, &adapter->dcb_cfg,
+ max_frame, DCB_TX_CONFIG);
+ ixgbe_dcb_calculate_tc_credits(&adapter->hw, &adapter->dcb_cfg,
+ max_frame, DCB_RX_CONFIG);
+
+ ixgbe_dcb_unpack_refill(&adapter->dcb_cfg,
+ DCB_TX_CONFIG, refill);
+ ixgbe_dcb_unpack_max(&adapter->dcb_cfg, max);
+ ixgbe_dcb_unpack_bwgid(&adapter->dcb_cfg,
+ DCB_TX_CONFIG, bwg_id);
+ ixgbe_dcb_unpack_prio(&adapter->dcb_cfg,
+ DCB_TX_CONFIG, prio_type);
+
+ ixgbe_dcb_hw_ets_config(&adapter->hw, refill, max,
+ bwg_id, prio_type);
+ }
+
if (adapter->dcb_cfg.pfc_mode_enable)
adapter->hw.fc.current_mode = ixgbe_fc_pfc;
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 12/16] ixgbe: dcb, use hardware independent routines
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: John Fastabend, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: John Fastabend <john.r.fastabend@intel.com>
This consolidates hardware specifics to ixgbe_dcb.c this simplifies
code that was previously branching based on hardware type.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_dcb_nl.c | 6 +-----
1 files changed, 1 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ixgbe/ixgbe_dcb_nl.c
index c94adec..a977df3 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_nl.c
@@ -405,11 +405,7 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
if (adapter->dcb_set_bitmap & BIT_PFC) {
u8 pfc_en;
ixgbe_dcb_unpack_pfc(&adapter->dcb_cfg, &pfc_en);
-
- if (adapter->hw.mac.type == ixgbe_mac_82598EB)
- ixgbe_dcb_config_pfc_82598(&adapter->hw, pfc_en);
- else if (adapter->hw.mac.type == ixgbe_mac_82599EB)
- ixgbe_dcb_config_pfc_82599(&adapter->hw, pfc_en);
+ ixgbe_dcb_hw_pfc_config(&adapter->hw, pfc_en);
ret = DCB_HW_CHG;
}
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 08/16] ixgbe: DCB, abstract out dcb_config from DCB hardware configuration
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: John Fastabend, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: John Fastabend <john.r.fastabend@intel.com>
Currently the routines that configure the HW for DCB require a
ixgbe_dcb_config structure. This structure was designed to support
the CEE standard and does not match the IEEE standard well.
This patch changes the HW routines in ixgbe_dcb_8259x.{ch} to use
raw pfc and bandwidth values. This requires some parsing of the DCB
configuration but makes the HW routines independent of the data
structure that contains the DCB configuration.
The primary advantage to doing this is we can do HW setup directly
from the 802.1Qaz ops without having to arbitrarily encapsulate this
data into the CEE structure.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_dcb.c | 74 ++++++++++++++++++++++-
drivers/net/ixgbe/ixgbe_dcb.h | 1 +
drivers/net/ixgbe/ixgbe_dcb_82598.c | 86 ++++++++++++++------------
drivers/net/ixgbe/ixgbe_dcb_82598.h | 23 ++++++-
drivers/net/ixgbe/ixgbe_dcb_82599.c | 115 ++++++++++++++++++++---------------
drivers/net/ixgbe/ixgbe_dcb_82599.h | 24 ++++++-
drivers/net/ixgbe/ixgbe_dcb_nl.c | 9 ++-
7 files changed, 230 insertions(+), 102 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_dcb.c b/drivers/net/ixgbe/ixgbe_dcb.c
index d16c260..d9bb670 100644
--- a/drivers/net/ixgbe/ixgbe_dcb.c
+++ b/drivers/net/ixgbe/ixgbe_dcb.c
@@ -141,6 +141,59 @@ out:
return ret_val;
}
+void ixgbe_dcb_unpack_pfc(struct ixgbe_dcb_config *cfg, u8 *pfc_en)
+{
+ int i;
+
+ *pfc_en = 0;
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++)
+ *pfc_en |= (cfg->tc_config[i].dcb_pfc & 0xF) << i;
+}
+
+void ixgbe_dcb_unpack_refill(struct ixgbe_dcb_config *cfg, int direction,
+ u16 *refill)
+{
+ struct tc_bw_alloc *p;
+ int i;
+
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ p = &cfg->tc_config[i].path[direction];
+ refill[i] = p->data_credits_refill;
+ }
+}
+
+void ixgbe_dcb_unpack_max(struct ixgbe_dcb_config *cfg, u16 *max)
+{
+ int i;
+
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++)
+ max[i] = cfg->tc_config[i].desc_credits_max;
+}
+
+void ixgbe_dcb_unpack_bwgid(struct ixgbe_dcb_config *cfg, int direction,
+ u8 *bwgid)
+{
+ struct tc_bw_alloc *p;
+ int i;
+
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ p = &cfg->tc_config[i].path[direction];
+ bwgid[i] = p->bwg_id;
+ }
+}
+
+void ixgbe_dcb_unpack_prio(struct ixgbe_dcb_config *cfg, int direction,
+ u8 *ptype)
+{
+ struct tc_bw_alloc *p;
+ int i;
+
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ p = &cfg->tc_config[i].path[direction];
+ ptype[i] = p->prio_type;
+ }
+}
+
/**
* ixgbe_dcb_hw_config - Config and enable DCB
* @hw: pointer to hardware structure
@@ -152,13 +205,30 @@ s32 ixgbe_dcb_hw_config(struct ixgbe_hw *hw,
struct ixgbe_dcb_config *dcb_config)
{
s32 ret = 0;
+ u8 pfc_en;
+ u8 ptype[MAX_TRAFFIC_CLASS];
+ u8 bwgid[MAX_TRAFFIC_CLASS];
+ u16 refill[MAX_TRAFFIC_CLASS];
+ u16 max[MAX_TRAFFIC_CLASS];
+
+ /* Unpack CEE standard containers */
+ ixgbe_dcb_unpack_pfc(dcb_config, &pfc_en);
+ ixgbe_dcb_unpack_refill(dcb_config, DCB_TX_CONFIG, refill);
+ ixgbe_dcb_unpack_max(dcb_config, max);
+ ixgbe_dcb_unpack_bwgid(dcb_config, DCB_TX_CONFIG, bwgid);
+ ixgbe_dcb_unpack_prio(dcb_config, DCB_TX_CONFIG, ptype);
+
switch (hw->mac.type) {
case ixgbe_mac_82598EB:
- ret = ixgbe_dcb_hw_config_82598(hw, dcb_config);
+ ret = ixgbe_dcb_hw_config_82598(hw, dcb_config->rx_pba_cfg,
+ pfc_en, refill, max, bwgid,
+ ptype);
break;
case ixgbe_mac_82599EB:
case ixgbe_mac_X540:
- ret = ixgbe_dcb_hw_config_82599(hw, dcb_config);
+ ret = ixgbe_dcb_hw_config_82599(hw, dcb_config->rx_pba_cfg,
+ pfc_en, refill, max, bwgid,
+ ptype);
break;
default:
break;
diff --git a/drivers/net/ixgbe/ixgbe_dcb.h b/drivers/net/ixgbe/ixgbe_dcb.h
index d0b2450..aa6cb5f 100644
--- a/drivers/net/ixgbe/ixgbe_dcb.h
+++ b/drivers/net/ixgbe/ixgbe_dcb.h
@@ -147,6 +147,7 @@ struct ixgbe_dcb_config {
};
/* DCB driver APIs */
+void ixgbe_dcb_unpack_pfc(struct ixgbe_dcb_config *cfg, u8 *pfc_en);
/* DCB credits calculation */
s32 ixgbe_dcb_calculate_tc_credits(struct ixgbe_hw *,
diff --git a/drivers/net/ixgbe/ixgbe_dcb_82598.c b/drivers/net/ixgbe/ixgbe_dcb_82598.c
index 19aa806..d128806 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_82598.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_82598.c
@@ -38,15 +38,14 @@
*
* Configure packet buffers for DCB mode.
*/
-static s32 ixgbe_dcb_config_packet_buffers_82598(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+static s32 ixgbe_dcb_config_packet_buffers_82598(struct ixgbe_hw *hw, u8 rx_pba)
{
s32 ret_val = 0;
u32 value = IXGBE_RXPBSIZE_64KB;
u8 i = 0;
/* Setup Rx packet buffer sizes */
- switch (dcb_config->rx_pba_cfg) {
+ switch (rx_pba) {
case pba_80_48:
/* Setup the first four at 80KB */
value = IXGBE_RXPBSIZE_80KB;
@@ -78,10 +77,11 @@ static s32 ixgbe_dcb_config_packet_buffers_82598(struct ixgbe_hw *hw,
*
* Configure Rx Data Arbiter and credits for each traffic class.
*/
-static s32 ixgbe_dcb_config_rx_arbiter_82598(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_rx_arbiter_82598(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *prio_type)
{
- struct tc_bw_alloc *p;
u32 reg = 0;
u32 credit_refill = 0;
u32 credit_max = 0;
@@ -102,13 +102,12 @@ static s32 ixgbe_dcb_config_rx_arbiter_82598(struct ixgbe_hw *hw,
/* Configure traffic class credits and priority */
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
- p = &dcb_config->tc_config[i].path[DCB_RX_CONFIG];
- credit_refill = p->data_credits_refill;
- credit_max = p->data_credits_max;
+ credit_refill = refill[i];
+ credit_max = max[i];
reg = credit_refill | (credit_max << IXGBE_RT2CR_MCL_SHIFT);
- if (p->prio_type == prio_link)
+ if (prio_type[i] == prio_link)
reg |= IXGBE_RT2CR_LSP;
IXGBE_WRITE_REG(hw, IXGBE_RT2CR(i), reg);
@@ -135,10 +134,12 @@ static s32 ixgbe_dcb_config_rx_arbiter_82598(struct ixgbe_hw *hw,
*
* Configure Tx Descriptor Arbiter and credits for each traffic class.
*/
-static s32 ixgbe_dcb_config_tx_desc_arbiter_82598(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_tx_desc_arbiter_82598(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type)
{
- struct tc_bw_alloc *p;
u32 reg, max_credits;
u8 i;
@@ -156,16 +157,15 @@ static s32 ixgbe_dcb_config_tx_desc_arbiter_82598(struct ixgbe_hw *hw,
/* Configure traffic class credits and priority */
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
- p = &dcb_config->tc_config[i].path[DCB_TX_CONFIG];
- max_credits = dcb_config->tc_config[i].desc_credits_max;
+ max_credits = max[i];
reg = max_credits << IXGBE_TDTQ2TCCR_MCL_SHIFT;
- reg |= p->data_credits_refill;
- reg |= (u32)(p->bwg_id) << IXGBE_TDTQ2TCCR_BWG_SHIFT;
+ reg |= refill[i];
+ reg |= (u32)(bwg_id[i]) << IXGBE_TDTQ2TCCR_BWG_SHIFT;
- if (p->prio_type == prio_group)
+ if (prio_type[i] == prio_group)
reg |= IXGBE_TDTQ2TCCR_GSP;
- if (p->prio_type == prio_link)
+ if (prio_type[i] == prio_link)
reg |= IXGBE_TDTQ2TCCR_LSP;
IXGBE_WRITE_REG(hw, IXGBE_TDTQ2TCCR(i), reg);
@@ -181,10 +181,12 @@ static s32 ixgbe_dcb_config_tx_desc_arbiter_82598(struct ixgbe_hw *hw,
*
* Configure Tx Data Arbiter and credits for each traffic class.
*/
-static s32 ixgbe_dcb_config_tx_data_arbiter_82598(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_tx_data_arbiter_82598(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type)
{
- struct tc_bw_alloc *p;
u32 reg;
u8 i;
@@ -198,15 +200,14 @@ static s32 ixgbe_dcb_config_tx_data_arbiter_82598(struct ixgbe_hw *hw,
/* Configure traffic class credits and priority */
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
- p = &dcb_config->tc_config[i].path[DCB_TX_CONFIG];
- reg = p->data_credits_refill;
- reg |= (u32)(p->data_credits_max) << IXGBE_TDPT2TCCR_MCL_SHIFT;
- reg |= (u32)(p->bwg_id) << IXGBE_TDPT2TCCR_BWG_SHIFT;
+ reg = refill[i];
+ reg |= (u32)(max[i]) << IXGBE_TDPT2TCCR_MCL_SHIFT;
+ reg |= (u32)(bwg_id[i]) << IXGBE_TDPT2TCCR_BWG_SHIFT;
- if (p->prio_type == prio_group)
+ if (prio_type[i] == prio_group)
reg |= IXGBE_TDPT2TCCR_GSP;
- if (p->prio_type == prio_link)
+ if (prio_type[i] == prio_link)
reg |= IXGBE_TDPT2TCCR_LSP;
IXGBE_WRITE_REG(hw, IXGBE_TDPT2TCCR(i), reg);
@@ -227,13 +228,12 @@ static s32 ixgbe_dcb_config_tx_data_arbiter_82598(struct ixgbe_hw *hw,
*
* Configure Priority Flow Control for each traffic class.
*/
-s32 ixgbe_dcb_config_pfc_82598(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_pfc_82598(struct ixgbe_hw *hw, u8 pfc_en)
{
u32 reg, rx_pba_size;
u8 i;
- if (!dcb_config->pfc_mode_enable)
+ if (!pfc_en)
goto out;
/* Enable Transmit Priority Flow Control */
@@ -254,19 +254,20 @@ s32 ixgbe_dcb_config_pfc_82598(struct ixgbe_hw *hw,
* for each traffic class.
*/
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ int enabled = pfc_en & (1 << i);
rx_pba_size = IXGBE_READ_REG(hw, IXGBE_RXPBSIZE(i));
rx_pba_size >>= IXGBE_RXPBSIZE_SHIFT;
reg = (rx_pba_size - hw->fc.low_water) << 10;
- if (dcb_config->tc_config[i].dcb_pfc == pfc_enabled_tx ||
- dcb_config->tc_config[i].dcb_pfc == pfc_enabled_full)
+ if (enabled == pfc_enabled_tx ||
+ enabled == pfc_enabled_full)
reg |= IXGBE_FCRTL_XONE;
IXGBE_WRITE_REG(hw, IXGBE_FCRTL(i), reg);
reg = (rx_pba_size - hw->fc.high_water) << 10;
- if (dcb_config->tc_config[i].dcb_pfc == pfc_enabled_tx ||
- dcb_config->tc_config[i].dcb_pfc == pfc_enabled_full)
+ if (enabled == pfc_enabled_tx ||
+ enabled == pfc_enabled_full)
reg |= IXGBE_FCRTH_FCEN;
IXGBE_WRITE_REG(hw, IXGBE_FCRTH(i), reg);
@@ -323,13 +324,16 @@ static s32 ixgbe_dcb_config_tc_stats_82598(struct ixgbe_hw *hw)
* Configure dcb settings and enable dcb mode.
*/
s32 ixgbe_dcb_hw_config_82598(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+ u8 rx_pba, u8 pfc_en, u16 *refill,
+ u16 *max, u8 *bwg_id, u8 *prio_type)
{
- ixgbe_dcb_config_packet_buffers_82598(hw, dcb_config);
- ixgbe_dcb_config_rx_arbiter_82598(hw, dcb_config);
- ixgbe_dcb_config_tx_desc_arbiter_82598(hw, dcb_config);
- ixgbe_dcb_config_tx_data_arbiter_82598(hw, dcb_config);
- ixgbe_dcb_config_pfc_82598(hw, dcb_config);
+ ixgbe_dcb_config_packet_buffers_82598(hw, rx_pba);
+ ixgbe_dcb_config_rx_arbiter_82598(hw, refill, max, prio_type);
+ ixgbe_dcb_config_tx_desc_arbiter_82598(hw, refill, max,
+ bwg_id, prio_type);
+ ixgbe_dcb_config_tx_data_arbiter_82598(hw, refill, max,
+ bwg_id, prio_type);
+ ixgbe_dcb_config_pfc_82598(hw, pfc_en);
ixgbe_dcb_config_tc_stats_82598(hw);
return 0;
diff --git a/drivers/net/ixgbe/ixgbe_dcb_82598.h b/drivers/net/ixgbe/ixgbe_dcb_82598.h
index abc03cc..0d2a758 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_82598.h
+++ b/drivers/net/ixgbe/ixgbe_dcb_82598.h
@@ -71,9 +71,28 @@
/* DCB hardware-specific driver APIs */
/* DCB PFC functions */
-s32 ixgbe_dcb_config_pfc_82598(struct ixgbe_hw *, struct ixgbe_dcb_config *);
+s32 ixgbe_dcb_config_pfc_82598(struct ixgbe_hw *, u8 pfc_en);
/* DCB hw initialization */
-s32 ixgbe_dcb_hw_config_82598(struct ixgbe_hw *, struct ixgbe_dcb_config *);
+s32 ixgbe_dcb_config_rx_arbiter_82598(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *prio_type);
+
+s32 ixgbe_dcb_config_tx_desc_arbiter_82598(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type);
+
+s32 ixgbe_dcb_config_tx_data_arbiter_82598(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type);
+
+s32 ixgbe_dcb_hw_config_82598(struct ixgbe_hw *hw,
+ u8 rx_pba, u8 pfc_en, u16 *refill,
+ u16 *max, u8 *bwg_id, u8 *prio_type);
#endif /* _DCB_82598_CONFIG_H */
diff --git a/drivers/net/ixgbe/ixgbe_dcb_82599.c b/drivers/net/ixgbe/ixgbe_dcb_82599.c
index 374e1f7..b0d97a9 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_82599.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_82599.c
@@ -33,19 +33,18 @@
/**
* ixgbe_dcb_config_packet_buffers_82599 - Configure DCB packet buffers
* @hw: pointer to hardware structure
- * @dcb_config: pointer to ixgbe_dcb_config structure
+ * @rx_pba: method to distribute packet buffer
*
* Configure packet buffers for DCB mode.
*/
-static s32 ixgbe_dcb_config_packet_buffers_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+static s32 ixgbe_dcb_config_packet_buffers_82599(struct ixgbe_hw *hw, u8 rx_pba)
{
s32 ret_val = 0;
u32 value = IXGBE_RXPBSIZE_64KB;
u8 i = 0;
/* Setup Rx packet buffer sizes */
- switch (dcb_config->rx_pba_cfg) {
+ switch (rx_pba) {
case pba_80_48:
/* Setup the first four at 80KB */
value = IXGBE_RXPBSIZE_80KB;
@@ -75,14 +74,19 @@ static s32 ixgbe_dcb_config_packet_buffers_82599(struct ixgbe_hw *hw,
/**
* ixgbe_dcb_config_rx_arbiter_82599 - Config Rx Data arbiter
* @hw: pointer to hardware structure
- * @dcb_config: pointer to ixgbe_dcb_config structure
+ * @refill: refill credits index by traffic class
+ * @max: max credits index by traffic class
+ * @bwg_id: bandwidth grouping indexed by traffic class
+ * @prio_type: priority type indexed by traffic class
*
* Configure Rx Packet Arbiter and credits for each traffic class.
*/
-static s32 ixgbe_dcb_config_rx_arbiter_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_rx_arbiter_82599(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type)
{
- struct tc_bw_alloc *p;
u32 reg = 0;
u32 credit_refill = 0;
u32 credit_max = 0;
@@ -103,15 +107,13 @@ static s32 ixgbe_dcb_config_rx_arbiter_82599(struct ixgbe_hw *hw,
/* Configure traffic class credits and priority */
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
- p = &dcb_config->tc_config[i].path[DCB_RX_CONFIG];
-
- credit_refill = p->data_credits_refill;
- credit_max = p->data_credits_max;
+ credit_refill = refill[i];
+ credit_max = max[i];
reg = credit_refill | (credit_max << IXGBE_RTRPT4C_MCL_SHIFT);
- reg |= (u32)(p->bwg_id) << IXGBE_RTRPT4C_BWG_SHIFT;
+ reg |= (u32)(bwg_id[i]) << IXGBE_RTRPT4C_BWG_SHIFT;
- if (p->prio_type == prio_link)
+ if (prio_type[i] == prio_link)
reg |= IXGBE_RTRPT4C_LSP;
IXGBE_WRITE_REG(hw, IXGBE_RTRPT4C(i), reg);
@@ -130,14 +132,19 @@ static s32 ixgbe_dcb_config_rx_arbiter_82599(struct ixgbe_hw *hw,
/**
* ixgbe_dcb_config_tx_desc_arbiter_82599 - Config Tx Desc. arbiter
* @hw: pointer to hardware structure
- * @dcb_config: pointer to ixgbe_dcb_config structure
+ * @refill: refill credits index by traffic class
+ * @max: max credits index by traffic class
+ * @bwg_id: bandwidth grouping indexed by traffic class
+ * @prio_type: priority type indexed by traffic class
*
* Configure Tx Descriptor Arbiter and credits for each traffic class.
*/
-static s32 ixgbe_dcb_config_tx_desc_arbiter_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_tx_desc_arbiter_82599(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type)
{
- struct tc_bw_alloc *p;
u32 reg, max_credits;
u8 i;
@@ -149,16 +156,15 @@ static s32 ixgbe_dcb_config_tx_desc_arbiter_82599(struct ixgbe_hw *hw,
/* Configure traffic class credits and priority */
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
- p = &dcb_config->tc_config[i].path[DCB_TX_CONFIG];
- max_credits = dcb_config->tc_config[i].desc_credits_max;
+ max_credits = max[i];
reg = max_credits << IXGBE_RTTDT2C_MCL_SHIFT;
- reg |= p->data_credits_refill;
- reg |= (u32)(p->bwg_id) << IXGBE_RTTDT2C_BWG_SHIFT;
+ reg |= refill[i];
+ reg |= (u32)(bwg_id[i]) << IXGBE_RTTDT2C_BWG_SHIFT;
- if (p->prio_type == prio_group)
+ if (prio_type[i] == prio_group)
reg |= IXGBE_RTTDT2C_GSP;
- if (p->prio_type == prio_link)
+ if (prio_type[i] == prio_link)
reg |= IXGBE_RTTDT2C_LSP;
IXGBE_WRITE_REG(hw, IXGBE_RTTDT2C(i), reg);
@@ -177,14 +183,19 @@ static s32 ixgbe_dcb_config_tx_desc_arbiter_82599(struct ixgbe_hw *hw,
/**
* ixgbe_dcb_config_tx_data_arbiter_82599 - Config Tx Data arbiter
* @hw: pointer to hardware structure
- * @dcb_config: pointer to ixgbe_dcb_config structure
+ * @refill: refill credits index by traffic class
+ * @max: max credits index by traffic class
+ * @bwg_id: bandwidth grouping indexed by traffic class
+ * @prio_type: priority type indexed by traffic class
*
* Configure Tx Packet Arbiter and credits for each traffic class.
*/
-static s32 ixgbe_dcb_config_tx_data_arbiter_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_tx_data_arbiter_82599(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type)
{
- struct tc_bw_alloc *p;
u32 reg;
u8 i;
@@ -205,15 +216,14 @@ static s32 ixgbe_dcb_config_tx_data_arbiter_82599(struct ixgbe_hw *hw,
/* Configure traffic class credits and priority */
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
- p = &dcb_config->tc_config[i].path[DCB_TX_CONFIG];
- reg = p->data_credits_refill;
- reg |= (u32)(p->data_credits_max) << IXGBE_RTTPT2C_MCL_SHIFT;
- reg |= (u32)(p->bwg_id) << IXGBE_RTTPT2C_BWG_SHIFT;
+ reg = refill[i];
+ reg |= (u32)(max[i]) << IXGBE_RTTPT2C_MCL_SHIFT;
+ reg |= (u32)(bwg_id[i]) << IXGBE_RTTPT2C_BWG_SHIFT;
- if (p->prio_type == prio_group)
+ if (prio_type[i] == prio_group)
reg |= IXGBE_RTTPT2C_GSP;
- if (p->prio_type == prio_link)
+ if (prio_type[i] == prio_link)
reg |= IXGBE_RTTPT2C_LSP;
IXGBE_WRITE_REG(hw, IXGBE_RTTPT2C(i), reg);
@@ -233,17 +243,16 @@ static s32 ixgbe_dcb_config_tx_data_arbiter_82599(struct ixgbe_hw *hw,
/**
* ixgbe_dcb_config_pfc_82599 - Configure priority flow control
* @hw: pointer to hardware structure
- * @dcb_config: pointer to ixgbe_dcb_config structure
+ * @pfc_en: enabled pfc bitmask
*
* Configure Priority Flow Control (PFC) for each traffic class.
*/
-s32 ixgbe_dcb_config_pfc_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+s32 ixgbe_dcb_config_pfc_82599(struct ixgbe_hw *hw, u8 pfc_en)
{
u32 i, reg, rx_pba_size;
/* If PFC is disabled globally then fall back to LFC. */
- if (!dcb_config->pfc_mode_enable) {
+ if (!pfc_en) {
for (i = 0; i < MAX_TRAFFIC_CLASS; i++)
hw->mac.ops.fc_enable(hw, i);
goto out;
@@ -251,19 +260,18 @@ s32 ixgbe_dcb_config_pfc_82599(struct ixgbe_hw *hw,
/* Configure PFC Tx thresholds per TC */
for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ int enabled = pfc_en & (1 << i);
rx_pba_size = IXGBE_READ_REG(hw, IXGBE_RXPBSIZE(i));
rx_pba_size >>= IXGBE_RXPBSIZE_SHIFT;
reg = (rx_pba_size - hw->fc.low_water) << 10;
- if (dcb_config->tc_config[i].dcb_pfc == pfc_enabled_full ||
- dcb_config->tc_config[i].dcb_pfc == pfc_enabled_tx)
+ if (enabled)
reg |= IXGBE_FCRTL_XONE;
IXGBE_WRITE_REG(hw, IXGBE_FCRTL_82599(i), reg);
reg = (rx_pba_size - hw->fc.high_water) << 10;
- if (dcb_config->tc_config[i].dcb_pfc == pfc_enabled_full ||
- dcb_config->tc_config[i].dcb_pfc == pfc_enabled_tx)
+ if (enabled)
reg |= IXGBE_FCRTH_FCEN;
IXGBE_WRITE_REG(hw, IXGBE_FCRTH_82599(i), reg);
}
@@ -349,7 +357,6 @@ static s32 ixgbe_dcb_config_tc_stats_82599(struct ixgbe_hw *hw)
/**
* ixgbe_dcb_config_82599 - Configure general DCB parameters
* @hw: pointer to hardware structure
- * @dcb_config: pointer to ixgbe_dcb_config structure
*
* Configure general DCB parameters.
*/
@@ -406,19 +413,27 @@ static s32 ixgbe_dcb_config_82599(struct ixgbe_hw *hw)
/**
* ixgbe_dcb_hw_config_82599 - Configure and enable DCB
* @hw: pointer to hardware structure
- * @dcb_config: pointer to ixgbe_dcb_config structure
+ * @rx_pba: method to distribute packet buffer
+ * @refill: refill credits index by traffic class
+ * @max: max credits index by traffic class
+ * @bwg_id: bandwidth grouping indexed by traffic class
+ * @prio_type: priority type indexed by traffic class
+ * @pfc_en: enabled pfc bitmask
*
* Configure dcb settings and enable dcb mode.
*/
s32 ixgbe_dcb_hw_config_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config)
+ u8 rx_pba, u8 pfc_en, u16 *refill,
+ u16 *max, u8 *bwg_id, u8 *prio_type)
{
- ixgbe_dcb_config_packet_buffers_82599(hw, dcb_config);
+ ixgbe_dcb_config_packet_buffers_82599(hw, rx_pba);
ixgbe_dcb_config_82599(hw);
- ixgbe_dcb_config_rx_arbiter_82599(hw, dcb_config);
- ixgbe_dcb_config_tx_desc_arbiter_82599(hw, dcb_config);
- ixgbe_dcb_config_tx_data_arbiter_82599(hw, dcb_config);
- ixgbe_dcb_config_pfc_82599(hw, dcb_config);
+ ixgbe_dcb_config_rx_arbiter_82599(hw, refill, max, bwg_id, prio_type);
+ ixgbe_dcb_config_tx_desc_arbiter_82599(hw, refill, max,
+ bwg_id, prio_type);
+ ixgbe_dcb_config_tx_data_arbiter_82599(hw, refill, max,
+ bwg_id, prio_type);
+ ixgbe_dcb_config_pfc_82599(hw, pfc_en);
ixgbe_dcb_config_tc_stats_82599(hw);
return 0;
diff --git a/drivers/net/ixgbe/ixgbe_dcb_82599.h b/drivers/net/ixgbe/ixgbe_dcb_82599.h
index 3841649..5b0ca85 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_82599.h
+++ b/drivers/net/ixgbe/ixgbe_dcb_82599.h
@@ -102,11 +102,29 @@
/* DCB hardware-specific driver APIs */
/* DCB PFC functions */
-s32 ixgbe_dcb_config_pfc_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *dcb_config);
+s32 ixgbe_dcb_config_pfc_82599(struct ixgbe_hw *hw, u8 pfc_en);
/* DCB hw initialization */
+s32 ixgbe_dcb_config_rx_arbiter_82599(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type);
+
+s32 ixgbe_dcb_config_tx_desc_arbiter_82599(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type);
+
+s32 ixgbe_dcb_config_tx_data_arbiter_82599(struct ixgbe_hw *hw,
+ u16 *refill,
+ u16 *max,
+ u8 *bwg_id,
+ u8 *prio_type);
+
s32 ixgbe_dcb_hw_config_82599(struct ixgbe_hw *hw,
- struct ixgbe_dcb_config *config);
+ u8 rx_pba, u8 pfc_en, u16 *refill,
+ u16 *max, u8 *bwg_id, u8 *prio_type);
#endif /* _DCB_82599_CONFIG_H */
diff --git a/drivers/net/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ixgbe/ixgbe_dcb_nl.c
index 4805835..6ab1f1a 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_nl.c
@@ -422,12 +422,13 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
}
ret = DCB_HW_CHG_RST;
} else if (adapter->dcb_set_bitmap & BIT_PFC) {
+ u8 pfc_en;
+ ixgbe_dcb_unpack_pfc(&adapter->dcb_cfg, &pfc_en);
+
if (adapter->hw.mac.type == ixgbe_mac_82598EB)
- ixgbe_dcb_config_pfc_82598(&adapter->hw,
- &adapter->dcb_cfg);
+ ixgbe_dcb_config_pfc_82598(&adapter->hw, pfc_en);
else if (adapter->hw.mac.type == ixgbe_mac_82599EB)
- ixgbe_dcb_config_pfc_82599(&adapter->hw,
- &adapter->dcb_cfg);
+ ixgbe_dcb_config_pfc_82599(&adapter->hw, pfc_en);
ret = DCB_HW_CHG;
}
if (adapter->dcb_cfg.pfc_mode_enable)
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 09/16] ixgbe: DCB, implement 802.1Qaz routines
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: John Fastabend, netdev, gospo, bphilips
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: John Fastabend <john.r.fastabend@intel.com>
Implements 802.1Qaz support for ixgbe driver. Additionally,
this adds IEEE_8021QAZ_TSA_{} defines to dcbnl.h this is to
avoid having to use cryptic numeric codes for the TSA type.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
---
drivers/net/ixgbe/ixgbe.h | 4 ++
drivers/net/ixgbe/ixgbe_dcb.c | 102 +++++++++++++++++++++++++++++++++++
drivers/net/ixgbe/ixgbe_dcb.h | 4 ++
drivers/net/ixgbe/ixgbe_dcb_82598.c | 2 +-
drivers/net/ixgbe/ixgbe_dcb_nl.c | 91 +++++++++++++++++++++++++++++++
include/linux/dcbnl.h | 5 ++
6 files changed, 207 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe.h b/drivers/net/ixgbe/ixgbe.h
index 3b8c924..d04afde 100644
--- a/drivers/net/ixgbe/ixgbe.h
+++ b/drivers/net/ixgbe/ixgbe.h
@@ -334,6 +334,10 @@ struct ixgbe_adapter {
u16 bd_number;
struct work_struct reset_task;
struct ixgbe_q_vector *q_vector[MAX_MSIX_Q_VECTORS];
+
+ /* DCB parameters */
+ struct ieee_pfc *ixgbe_ieee_pfc;
+ struct ieee_ets *ixgbe_ieee_ets;
struct ixgbe_dcb_config dcb_cfg;
struct ixgbe_dcb_config temp_dcb_cfg;
u8 dcb_set_bitmap;
diff --git a/drivers/net/ixgbe/ixgbe_dcb.c b/drivers/net/ixgbe/ixgbe_dcb.c
index d9bb670..0d7830f 100644
--- a/drivers/net/ixgbe/ixgbe_dcb.c
+++ b/drivers/net/ixgbe/ixgbe_dcb.c
@@ -34,6 +34,41 @@
#include "ixgbe_dcb_82599.h"
/**
+ * ixgbe_ieee_credits - This calculates the ieee traffic class
+ * credits from the configured bandwidth percentages. Credits
+ * are the smallest unit programable into the underlying
+ * hardware. The IEEE 802.1Qaz specification do not use bandwidth
+ * groups so this is much simplified from the CEE case.
+ */
+s32 ixgbe_ieee_credits(__u8 *bw, __u16 *refill, __u16 *max, int max_frame)
+{
+ int min_percent = 100;
+ int min_credit, multiplier;
+ int i;
+
+ min_credit = ((max_frame / 2) + DCB_CREDIT_QUANTUM - 1) /
+ DCB_CREDIT_QUANTUM;
+
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ if (bw[i] < min_percent && bw[i])
+ min_percent = bw[i];
+ }
+
+ multiplier = (min_credit / min_percent) + 1;
+
+ /* Find out the hw credits for each TC */
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ int val = min(bw[i] * multiplier, MAX_CREDIT_REFILL);
+ if (val < min_credit)
+ val = min_credit;
+ refill[i] = val;
+
+ max[i] = (bw[i] * MAX_CREDIT)/100;
+ }
+ return 0;
+}
+
+/**
* ixgbe_dcb_calculate_tc_credits - Calculates traffic class credits
* @ixgbe_dcb_config: Struct containing DCB settings.
* @direction: Configuring either Tx or Rx.
@@ -236,3 +271,70 @@ s32 ixgbe_dcb_hw_config(struct ixgbe_hw *hw,
return ret;
}
+/* Helper routines to abstract HW specifics from DCB netlink ops */
+s32 ixgbe_dcb_hw_pfc_config(struct ixgbe_hw *hw, u8 pfc_en)
+{
+ int ret = -EINVAL;
+
+ switch (hw->mac.type) {
+ case ixgbe_mac_82598EB:
+ ret = ixgbe_dcb_config_pfc_82598(hw, pfc_en);
+ break;
+ case ixgbe_mac_82599EB:
+ case ixgbe_mac_X540:
+ ret = ixgbe_dcb_config_pfc_82599(hw, pfc_en);
+ break;
+ default:
+ break;
+ }
+ return ret;
+}
+
+s32 ixgbe_dcb_hw_ets_config(struct ixgbe_hw *hw,
+ u16 *refill, u16 *max, u8 *bwg_id, u8 *tsa)
+{
+ int i;
+ u8 prio_type[IEEE_8021QAZ_MAX_TCS];
+
+ /* Map TSA onto CEE prio type */
+ for (i = 0; i < IEEE_8021QAZ_MAX_TCS; i++) {
+ switch (tsa[i]) {
+ case IEEE_8021QAZ_TSA_STRICT:
+ prio_type[i] = 2;
+ break;
+ case IEEE_8021QAZ_TSA_ETS:
+ prio_type[i] = 0;
+ break;
+ default:
+ /* Hardware only supports priority strict or
+ * ETS transmission selection algorithms if
+ * we receive some other value from dcbnl
+ * throw an error
+ */
+ return -EINVAL;
+ }
+ }
+
+ switch (hw->mac.type) {
+ case ixgbe_mac_82598EB:
+ ixgbe_dcb_config_rx_arbiter_82598(hw, refill, max,
+ prio_type);
+ ixgbe_dcb_config_tx_desc_arbiter_82598(hw, refill, max,
+ bwg_id, prio_type);
+ ixgbe_dcb_config_tx_data_arbiter_82598(hw, refill, max,
+ bwg_id, prio_type);
+ break;
+ case ixgbe_mac_82599EB:
+ case ixgbe_mac_X540:
+ ixgbe_dcb_config_rx_arbiter_82599(hw, refill, max,
+ bwg_id, prio_type);
+ ixgbe_dcb_config_tx_desc_arbiter_82599(hw, refill, max,
+ bwg_id, prio_type);
+ ixgbe_dcb_config_tx_data_arbiter_82599(hw, refill, max,
+ bwg_id, prio_type);
+ break;
+ default:
+ break;
+ }
+ return 0;
+}
diff --git a/drivers/net/ixgbe/ixgbe_dcb.h b/drivers/net/ixgbe/ixgbe_dcb.h
index aa6cb5f..4e4a641 100644
--- a/drivers/net/ixgbe/ixgbe_dcb.h
+++ b/drivers/net/ixgbe/ixgbe_dcb.h
@@ -150,10 +150,14 @@ struct ixgbe_dcb_config {
void ixgbe_dcb_unpack_pfc(struct ixgbe_dcb_config *cfg, u8 *pfc_en);
/* DCB credits calculation */
+s32 ixgbe_ieee_credits(__u8 *bw, __u16 *refill, __u16 *max, int max_frame);
s32 ixgbe_dcb_calculate_tc_credits(struct ixgbe_hw *,
struct ixgbe_dcb_config *, int, u8);
/* DCB hw initialization */
+s32 ixgbe_dcb_hw_ets_config(struct ixgbe_hw *hw,
+ u16 *refill, u16 *max, u8 *bwg_id, u8 *prio_type);
+s32 ixgbe_dcb_hw_pfc_config(struct ixgbe_hw *hw, u8 pfc_en);
s32 ixgbe_dcb_hw_config(struct ixgbe_hw *, struct ixgbe_dcb_config *);
/* DCB definitions for credit calculation */
diff --git a/drivers/net/ixgbe/ixgbe_dcb_82598.c b/drivers/net/ixgbe/ixgbe_dcb_82598.c
index d128806..2965edc 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_82598.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_82598.c
@@ -291,7 +291,7 @@ out:
* Configure queue statistics registers, all queues belonging to same traffic
* class uses a single set of queue statistics counters.
*/
-static s32 ixgbe_dcb_config_tc_stats_82598(struct ixgbe_hw *hw)
+s32 ixgbe_dcb_config_tc_stats_82598(struct ixgbe_hw *hw)
{
u32 reg = 0;
u8 i = 0;
diff --git a/drivers/net/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ixgbe/ixgbe_dcb_nl.c
index 6ab1f1a..e75a3c9 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_nl.c
@@ -606,7 +606,98 @@ static u8 ixgbe_dcbnl_setapp(struct net_device *netdev,
return rval;
}
+static int ixgbe_dcbnl_ieee_getets(struct net_device *dev,
+ struct ieee_ets *ets)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(dev);
+ struct ieee_ets *my_ets = adapter->ixgbe_ieee_ets;
+
+ /* No IEEE PFC settings available */
+ if (!my_ets)
+ return -EINVAL;
+
+ ets->ets_cap = MAX_TRAFFIC_CLASS;
+ ets->cbs = my_ets->cbs;
+ memcpy(ets->tc_tx_bw, my_ets->tc_tx_bw, sizeof(ets->tc_tx_bw));
+ memcpy(ets->tc_rx_bw, my_ets->tc_rx_bw, sizeof(ets->tc_rx_bw));
+ memcpy(ets->tc_tsa, my_ets->tc_tsa, sizeof(ets->tc_tsa));
+ memcpy(ets->prio_tc, my_ets->prio_tc, sizeof(ets->prio_tc));
+ return 0;
+}
+
+static int ixgbe_dcbnl_ieee_setets(struct net_device *dev,
+ struct ieee_ets *ets)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(dev);
+ __u16 refill[IEEE_8021QAZ_MAX_TCS], max[IEEE_8021QAZ_MAX_TCS];
+ int max_frame = dev->mtu + ETH_HLEN + ETH_FCS_LEN;
+ int err;
+ /* naively give each TC a bwg to map onto CEE hardware */
+ __u8 bwg_id[IEEE_8021QAZ_MAX_TCS] = {0, 1, 2, 3, 4, 5, 6, 7};
+
+ if (!adapter->ixgbe_ieee_ets) {
+ adapter->ixgbe_ieee_ets = kmalloc(sizeof(struct ieee_ets),
+ GFP_KERNEL);
+ if (!adapter->ixgbe_ieee_ets)
+ return -ENOMEM;
+ }
+
+
+ memcpy(adapter->ixgbe_ieee_ets, ets, sizeof(*adapter->ixgbe_ieee_ets));
+
+ ixgbe_ieee_credits(ets->tc_tx_bw, refill, max, max_frame);
+ err = ixgbe_dcb_hw_ets_config(&adapter->hw, refill, max,
+ bwg_id, ets->tc_tsa);
+ return err;
+}
+
+static int ixgbe_dcbnl_ieee_getpfc(struct net_device *dev,
+ struct ieee_pfc *pfc)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(dev);
+ struct ieee_pfc *my_pfc = adapter->ixgbe_ieee_pfc;
+ int i;
+
+ /* No IEEE PFC settings available */
+ if (!my_pfc)
+ return -EINVAL;
+
+ pfc->pfc_cap = MAX_TRAFFIC_CLASS;
+ pfc->pfc_en = my_pfc->pfc_en;
+ pfc->mbc = my_pfc->mbc;
+ pfc->delay = my_pfc->delay;
+
+ for (i = 0; i < MAX_TRAFFIC_CLASS; i++) {
+ pfc->requests[i] = adapter->stats.pxoffrxc[i];
+ pfc->indications[i] = adapter->stats.pxofftxc[i];
+ }
+
+ return 0;
+}
+
+static int ixgbe_dcbnl_ieee_setpfc(struct net_device *dev,
+ struct ieee_pfc *pfc)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(dev);
+ int err;
+
+ if (!adapter->ixgbe_ieee_pfc) {
+ adapter->ixgbe_ieee_pfc = kmalloc(sizeof(struct ieee_pfc),
+ GFP_KERNEL);
+ if (!adapter->ixgbe_ieee_pfc)
+ return -ENOMEM;
+ }
+
+ memcpy(adapter->ixgbe_ieee_pfc, pfc, sizeof(*adapter->ixgbe_ieee_pfc));
+ err = ixgbe_dcb_hw_pfc_config(&adapter->hw, pfc->pfc_en);
+ return err;
+}
+
const struct dcbnl_rtnl_ops dcbnl_ops = {
+ .ieee_getets = ixgbe_dcbnl_ieee_getets,
+ .ieee_setets = ixgbe_dcbnl_ieee_setets,
+ .ieee_getpfc = ixgbe_dcbnl_ieee_getpfc,
+ .ieee_setpfc = ixgbe_dcbnl_ieee_setpfc,
.getstate = ixgbe_dcbnl_get_state,
.setstate = ixgbe_dcbnl_set_state,
.getpermhwaddr = ixgbe_dcbnl_get_perm_hw_addr,
diff --git a/include/linux/dcbnl.h b/include/linux/dcbnl.h
index 68cd248..cd8d518 100644
--- a/include/linux/dcbnl.h
+++ b/include/linux/dcbnl.h
@@ -25,6 +25,11 @@
/* IEEE 802.1Qaz std supported values */
#define IEEE_8021QAZ_MAX_TCS 8
+#define IEEE_8021QAZ_TSA_STRICT 0
+#define IEEE_8021QAZ_TSA_CB_SHABER 1
+#define IEEE_8021QAZ_TSA_ETS 2
+#define IEEE_8021QAZ_TSA_VENDOR 255
+
/* This structure contains the IEEE 802.1Qaz ETS managed object
*
* @willing: willing bit in ETS configuratin TLV
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 07/16] ixgbe: DCB, remove round robin mode on 82598 devices
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: John Fastabend, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: John Fastabend <john.r.fastabend@intel.com>
Remove round robin configuration code for 82598 parts it
is not settable and is always false.
If we need/want this in the future we can add it back properly.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_dcb.h | 1 -
drivers/net/ixgbe/ixgbe_dcb_82598.c | 6 ++----
drivers/net/ixgbe/ixgbe_main.c | 1 -
3 files changed, 2 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_dcb.h b/drivers/net/ixgbe/ixgbe_dcb.h
index 1cfe38e..d0b2450 100644
--- a/drivers/net/ixgbe/ixgbe_dcb.h
+++ b/drivers/net/ixgbe/ixgbe_dcb.h
@@ -139,7 +139,6 @@ struct ixgbe_dcb_config {
struct tc_configuration tc_config[MAX_TRAFFIC_CLASS];
u8 bw_percentage[2][MAX_BW_GROUP]; /* One each for Tx/Rx */
bool pfc_mode_enable;
- bool round_robin_enable;
enum dcb_rx_pba_cfg rx_pba_cfg;
diff --git a/drivers/net/ixgbe/ixgbe_dcb_82598.c b/drivers/net/ixgbe/ixgbe_dcb_82598.c
index 9a5e89c..19aa806 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_82598.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_82598.c
@@ -146,10 +146,8 @@ static s32 ixgbe_dcb_config_tx_desc_arbiter_82598(struct ixgbe_hw *hw,
/* Enable arbiter */
reg &= ~IXGBE_DPMCS_ARBDIS;
- if (!(dcb_config->round_robin_enable)) {
- /* Enable DFP and Recycle mode */
- reg |= (IXGBE_DPMCS_TDPAC | IXGBE_DPMCS_TRM);
- }
+ /* Enable DFP and Recycle mode */
+ reg |= (IXGBE_DPMCS_TDPAC | IXGBE_DPMCS_TRM);
reg |= IXGBE_DPMCS_TSOEF;
/* Configure Max TSO packet size 34KB including payload and headers */
reg |= (0x4 << IXGBE_DPMCS_MTSOS_SHIFT);
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 602078b..ef4a517 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -5169,7 +5169,6 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter)
adapter->dcb_cfg.bw_percentage[DCB_RX_CONFIG][0] = 100;
adapter->dcb_cfg.rx_pba_cfg = pba_equal;
adapter->dcb_cfg.pfc_mode_enable = false;
- adapter->dcb_cfg.round_robin_enable = false;
adapter->dcb_set_bitmap = 0x00;
ixgbe_copy_dcb_cfg(&adapter->dcb_cfg, &adapter->temp_dcb_cfg,
adapter->ring_feature[RING_F_DCB].indices);
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 05/16] igb: Enable PF side of SR-IOV support for i350 devices
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: Carolyn Wyborny, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
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());
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 06/16] ixgbe: DCB, only reprogram HW if the FCoE priority is changed
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: John Fastabend, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: John Fastabend <john.r.fastabend@intel.com>
If the FCoE priority is not changing do not set the RESET and
APP_UPCHG bits. This causes unneeded HW resets and which can
cause unneeded LLDP frames and negotiations.
The current check is not sufficient because the FCoE priority
can change twice during a negotiation which results in the
bits being set. This occurs when the switch changes the
priority or when the link is reset with switches that do not
include the APP priority until after PFC has been negotiated.
This results in set_app being called with the local APP
priority. Then the negotiation completes and set_app
is called again with the peer APP priority. The check
fails so the device is reset and the above occurs again
resulting in an endless loop of resets.
By only resetting the device if the APP priority has really
changed we short circuit the loop.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Tested-by: Ross Brattain <ross.b.brattain@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_dcb_nl.c | 38 ++++++++++++++++++++++++++------------
1 files changed, 26 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_dcb_nl.c b/drivers/net/ixgbe/ixgbe_dcb_nl.c
index bf566e8..4805835 100644
--- a/drivers/net/ixgbe/ixgbe_dcb_nl.c
+++ b/drivers/net/ixgbe/ixgbe_dcb_nl.c
@@ -353,6 +353,7 @@ static void ixgbe_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority,
static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
{
struct ixgbe_adapter *adapter = netdev_priv(netdev);
+ bool do_reset;
int ret;
if (!adapter->dcb_set_bitmap)
@@ -368,7 +369,9 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
* Only take down the adapter if the configuration change
* requires a reset.
*/
- if (adapter->dcb_set_bitmap & BIT_RESETLINK) {
+ do_reset = adapter->dcb_set_bitmap & (BIT_RESETLINK | BIT_APP_UPCHG);
+
+ if (do_reset) {
while (test_and_set_bit(__IXGBE_RESETTING, &adapter->state))
msleep(1);
@@ -408,7 +411,7 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
}
}
- if (adapter->dcb_set_bitmap & BIT_RESETLINK) {
+ if (do_reset) {
if (adapter->dcb_set_bitmap & BIT_APP_UPCHG) {
ixgbe_init_interrupt_scheme(adapter);
if (netif_running(netdev))
@@ -430,7 +433,7 @@ static u8 ixgbe_dcbnl_set_all(struct net_device *netdev)
if (adapter->dcb_cfg.pfc_mode_enable)
adapter->hw.fc.current_mode = ixgbe_fc_pfc;
- if (adapter->dcb_set_bitmap & BIT_RESETLINK)
+ if (do_reset)
clear_bit(__IXGBE_RESETTING, &adapter->state);
adapter->dcb_set_bitmap = 0x00;
return ret;
@@ -568,18 +571,29 @@ static u8 ixgbe_dcbnl_setapp(struct net_device *netdev,
case DCB_APP_IDTYPE_ETHTYPE:
#ifdef IXGBE_FCOE
if (id == ETH_P_FCOE) {
- u8 tc;
- struct ixgbe_adapter *adapter;
+ u8 old_tc;
+ struct ixgbe_adapter *adapter = netdev_priv(netdev);
- adapter = netdev_priv(netdev);
- tc = adapter->fcoe.tc;
+ /* Get current programmed tc */
+ old_tc = adapter->fcoe.tc;
rval = ixgbe_fcoe_setapp(adapter, up);
- if ((!rval) && (tc != adapter->fcoe.tc) &&
- (adapter->flags & IXGBE_FLAG_DCB_ENABLED) &&
- (adapter->flags & IXGBE_FLAG_FCOE_ENABLED)) {
+
+ if (rval ||
+ !(adapter->flags & IXGBE_FLAG_DCB_ENABLED) ||
+ !(adapter->flags & IXGBE_FLAG_FCOE_ENABLED))
+ break;
+
+ /* The FCoE application priority may be changed multiple
+ * times in quick sucession with switches that build up
+ * TLVs. To avoid creating uneeded device resets this
+ * checks the actual HW configuration and clears
+ * BIT_APP_UPCHG if a HW configuration change is not
+ * need
+ */
+ if (old_tc == adapter->fcoe.tc)
+ adapter->dcb_set_bitmap &= ~BIT_APP_UPCHG;
+ else
adapter->dcb_set_bitmap |= BIT_APP_UPCHG;
- adapter->dcb_set_bitmap |= BIT_RESETLINK;
- }
}
#endif
break;
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 04/16] e1000e: return appropriate errors for 'ethtool -r'
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: Bruce Allan, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Bruce Allan <bruce.w.allan@intel.com>
...when invoked while interface is not up or when auto-negotiation is
disabled as done by other drivers.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/e1000e/ethtool.c | 11 +++++++++--
1 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/net/e1000e/ethtool.c b/drivers/net/e1000e/ethtool.c
index 0c08599..65ef9b5 100644
--- a/drivers/net/e1000e/ethtool.c
+++ b/drivers/net/e1000e/ethtool.c
@@ -1963,8 +1963,15 @@ static int e1000_set_coalesce(struct net_device *netdev,
static int e1000_nway_reset(struct net_device *netdev)
{
struct e1000_adapter *adapter = netdev_priv(netdev);
- if (netif_running(netdev))
- e1000e_reinit_locked(adapter);
+
+ if (!netif_running(netdev))
+ return -EAGAIN;
+
+ if (!adapter->hw.mac.autoneg)
+ return -EINVAL;
+
+ e1000e_reinit_locked(adapter);
+
return 0;
}
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 02/16] e1000e: use correct pointer when memcpy'ing a 2-dimensional array
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: Bruce Allan, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Bruce Allan <bruce.w.allan@intel.com>
*e1000_gstrings_test is not the same size as e1000_gstrings_test.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/e1000e/ethtool.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/e1000e/ethtool.c b/drivers/net/e1000e/ethtool.c
index daa7fe4..0c08599 100644
--- a/drivers/net/e1000e/ethtool.c
+++ b/drivers/net/e1000e/ethtool.c
@@ -2006,7 +2006,7 @@ static void e1000_get_strings(struct net_device *netdev, u32 stringset,
switch (stringset) {
case ETH_SS_TEST:
- memcpy(data, *e1000_gstrings_test, sizeof(e1000_gstrings_test));
+ memcpy(data, e1000_gstrings_test, sizeof(e1000_gstrings_test));
break;
case ETH_SS_STATS:
for (i = 0; i < E1000_GLOBAL_STATS_LEN; i++) {
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 03/16] e1000e: do not wakeup Tx queue until ready
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: Bruce Allan, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Bruce Allan <bruce.w.allan@intel.com>
When restarting the adapter via calls to e1000e_down() followed by
e1000e_up() (for example when restarting autonegotiation via ethtool),
packets can be queued for transmit before link is actually up and
netif_carrier_on is set. This was causing the watchdog_task to do an
extra unnecessary reset of the adapter. Delaying the wakeup of the Tx
queue until link is up prevents the reset.
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/e1000e/netdev.c | 5 +----
1 files changed, 1 insertions(+), 4 deletions(-)
diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 80614c9..3c95d6b 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -3324,8 +3324,6 @@ int e1000e_up(struct e1000_adapter *adapter)
e1000_configure_msix(adapter);
e1000_irq_enable(adapter);
- netif_wake_queue(adapter->netdev);
-
/* fire a link change interrupt to start the watchdog */
if (adapter->msix_entries)
ew32(ICS, E1000_ICS_LSC | E1000_ICR_OTHER);
@@ -3638,8 +3636,6 @@ static int e1000_open(struct net_device *netdev)
e1000_irq_enable(adapter);
- netif_start_queue(netdev);
-
adapter->idle_check = true;
pm_runtime_put(&pdev->dev);
@@ -4265,6 +4261,7 @@ static void e1000_watchdog_task(struct work_struct *work)
phy->ops.cfg_on_link_up(hw);
netif_carrier_on(netdev);
+ netif_wake_queue(netdev);
if (!test_bit(__E1000_DOWN, &adapter->state))
mod_timer(&adapter->phy_info_timer,
--
1.7.4
^ permalink raw reply related
* [net-next-2.6 00/16][pull request] Intel Wired LAN Driver Updates
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: Jeff Kirsher, netdev, gospo, bphilips
The following series contains e1000e cleanup patches, the addition of igb
PF support for i350 devices and several ixgbe cleanup patches.
The following are changes since commit 2360d2e8f01043632d6b651672bec66c49892f94:
enic: Update MAINTAINERS
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
Bruce Allan (4):
e1000e: replace unbounded sprintf with snprintf
e1000e: use correct pointer when memcpy'ing a 2-dimensional array
e1000e: do not wakeup Tx queue until ready
e1000e: return appropriate errors for 'ethtool -r'
Carolyn Wyborny (1):
igb: Enable PF side of SR-IOV support for i350 devices
Don Skidmore (3):
ixgbe: fix namespace issue with ixgbe_dcb_txq_to_tc
ixgbe: cleanup namespace complaint by removing little used function
ixgbe: cleanup ixgbe_init_mbx_params_pf namespace issue
John Fastabend (7):
ixgbe: DCB, only reprogram HW if the FCoE priority is changed
ixgbe: DCB, remove round robin mode on 82598 devices
ixgbe: DCB, abstract out dcb_config from DCB hardware configuration
ixgbe: DCB, implement 802.1Qaz routines
ixgbe: DCB, do not reset on CEE pg changes
ixgbe: DCB, remove RESET bit it is no longer needed
ixgbe: dcb, use hardware independent routines
drivers/net/e1000e/ethtool.c | 13 ++-
drivers/net/e1000e/netdev.c | 13 +-
drivers/net/igb/e1000_82575.c | 10 ++-
drivers/net/igb/e1000_mbx.c | 38 +++----
drivers/net/igb/igb_main.c | 9 +-
drivers/net/ixgbe/ixgbe.h | 5 +-
drivers/net/ixgbe/ixgbe_common.h | 4 +-
drivers/net/ixgbe/ixgbe_dcb.c | 176 +++++++++++++++++++++++++++++-
drivers/net/ixgbe/ixgbe_dcb.h | 10 ++-
drivers/net/ixgbe/ixgbe_dcb_82598.c | 94 ++++++++--------
drivers/net/ixgbe/ixgbe_dcb_82598.h | 23 ++++-
drivers/net/ixgbe/ixgbe_dcb_82599.c | 115 +++++++++++--------
drivers/net/ixgbe/ixgbe_dcb_82599.h | 24 ++++-
drivers/net/ixgbe/ixgbe_dcb_nl.c | 211 ++++++++++++++++++++++++++--------
drivers/net/ixgbe/ixgbe_ethtool.c | 34 ++++++-
drivers/net/ixgbe/ixgbe_main.c | 18 +--
drivers/net/ixgbe/ixgbe_mbx.c | 2 +
drivers/net/ixgbe/ixgbe_mbx.h | 2 +
include/linux/dcbnl.h | 5 +
19 files changed, 601 insertions(+), 205 deletions(-)
--
1.7.4
^ permalink raw reply
* [net-next-2.6 01/16] e1000e: replace unbounded sprintf with snprintf
From: Jeff Kirsher @ 2011-02-08 12:29 UTC (permalink / raw)
To: davem; +Cc: Bruce Allan, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297168167-15755-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Bruce Allan <bruce.w.allan@intel.com>
Signed-off-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/e1000e/netdev.c | 8 ++++++--
1 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 5b916b0..80614c9 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -1843,7 +1843,9 @@ static int e1000_request_msix(struct e1000_adapter *adapter)
int err = 0, vector = 0;
if (strlen(netdev->name) < (IFNAMSIZ - 5))
- sprintf(adapter->rx_ring->name, "%s-rx-0", netdev->name);
+ snprintf(adapter->rx_ring->name,
+ sizeof(adapter->rx_ring->name) - 1,
+ "%s-rx-0", netdev->name);
else
memcpy(adapter->rx_ring->name, netdev->name, IFNAMSIZ);
err = request_irq(adapter->msix_entries[vector].vector,
@@ -1856,7 +1858,9 @@ static int e1000_request_msix(struct e1000_adapter *adapter)
vector++;
if (strlen(netdev->name) < (IFNAMSIZ - 5))
- sprintf(adapter->tx_ring->name, "%s-tx-0", netdev->name);
+ snprintf(adapter->tx_ring->name,
+ sizeof(adapter->tx_ring->name) - 1,
+ "%s-tx-0", netdev->name);
else
memcpy(adapter->tx_ring->name, netdev->name, IFNAMSIZ);
err = request_irq(adapter->msix_entries[vector].vector,
--
1.7.4
^ permalink raw reply related
* RE: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support for Bosch C_CAN controller
From: Bhupesh SHARMA @ 2011-02-08 11:36 UTC (permalink / raw)
To: Wolfgang Grandegger
Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org,
netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Marc Kleine-Budde
In-Reply-To: <4D5128C8.1060208-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
> -----Original Message-----
> From: Wolfgang Grandegger [mailto:wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org]
> Sent: Tuesday, February 08, 2011 4:58 PM
> To: Bhupesh SHARMA
> Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org; netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; Marc
> Kleine-Budde
> Subject: Re: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support for
> Bosch C_CAN controller
>
> On 02/08/2011 11:45 AM, Bhupesh SHARMA wrote:
> >> -----Original Message-----
> >> From: Wolfgang Grandegger [mailto:wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org]
> >> Sent: Tuesday, February 08, 2011 4:01 PM
> >> To: Bhupesh SHARMA
> >> Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org; netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; Marc
> >> Kleine-Budde
> >> Subject: Re: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support
> for
> >> Bosch C_CAN controller
> >>
> >> On 02/08/2011 10:04 AM, Bhupesh SHARMA wrote:
> >>> Hi Wolfgang,
> >>>
> >>>>> + stats->rx_errors++;
> >>>>> + cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
> >>>>> + cf->data[2] |= CAN_ERR_PROT_UNSPEC;
> >>>>> +
> >>>>> + switch (lec_type) {
> >>>>> + case LEC_STUFF_ERROR:
> >>>>> + netdev_dbg(dev, "stuff error\n");
> >>>>> + cf->data[2] |= CAN_ERR_PROT_STUFF;
> >>>>> + break;
> >>>>> +
> >>>>> + case LEC_FORM_ERROR:
> >>>>> + netdev_dbg(dev, "form error\n");
> >>>>> + cf->data[2] |= CAN_ERR_PROT_FORM;
> >>>>> + break;
> >>>>> +
> >>>>> + case LEC_ACK_ERROR:
> >>>>> + netdev_dbg(dev, "ack error\n");
> >>>>> + cf->data[2] |= (CAN_ERR_PROT_LOC_ACK |
> >>>>> + CAN_ERR_PROT_LOC_ACK_DEL);
> >>>>> + break;
> >>>>> +
> >>>>> + case LEC_BIT1_ERROR:
> >>>>> + netdev_dbg(dev, "bit1 error\n");
> >>>>> + cf->data[2] |= CAN_ERR_PROT_BIT1;
> >>>>> + break;
> >>>>> +
> >>>>> + case LEC_BIT0_ERROR:
> >>>>> + netdev_dbg(dev, "bit0 error\n");
> >>>>> + cf->data[2] |= CAN_ERR_PROT_BIT0;
> >>>>> + break;
> >>>>> +
> >>>>> + case LEC_CRC_ERROR:
> >>>>> + netdev_dbg(dev, "CRC error\n");
> >>>>> + cf->data[2] |= (CAN_ERR_PROT_LOC_CRC_SEQ |
> >>>>> + CAN_ERR_PROT_LOC_CRC_DEL);
> >>>>> + break;
> >>>>> + }
> >>>
> >>> >From the C_CAN manual:
> >>>>
> >>>> "The LEC field holds a code which indicates the type of the last
> >> error
> >>>> to occur on the CAN bus. This field will be cleared to '0' when a
> >>>> message has been transferred (reception or transmission) without
> >> error.
> >>>> The unused code '7' may be written by the CPU to check for
> updates."
> >>>
> >>>> Not sure if it's necessary to reset the lec at init and after an
> >> error
> >>>> to 0x7 and check it. More below...
> >>>
> >>> I worked on your suggestion and instead found that the follow algo
> >> must be used
> >>> for reading updated `lec` values:
> >>> a. Init lec by 0x07 at start.
> >>> b. In function `c_can_err` check if lec is 0x7 (no bus error since
> >> this value was
> >>> written by CPU on status register) or 0x0 (no error). If so, return
> >> without
> >>> sending an error frame on stack. Else, check for the lec error type
> >> and
> >>> submit error frame on stack accordingly.
> >>> c. In case a lec error is found and served in `c_can_err` routine,
> >> write
> >>> lec value to 0x07 again in status reg to check for updated lec
> later.
> >>>
> >>> This is my understanding after reading the specs time and again and
> >>> implementing/testing the logic.
> >>>
> >>> Do you think this is fine or do you have any better approach?
> >>
> >> That's what I remember from the CC770 driver. Search for lec in:
> >>
> >>
> http://svn.berlios.de/wsvn/socketcan/trunk/kernel/2.6/drivers/net/can/c
> >> c770/cc770.c
> >>
> >
> > Seems similar. But step (c) mentioned above seems missing from
> cc770.c driver,
> > i.e. "In case a lec error is found and served (by means of sending an
> error
> > frame on the bus) write lec value to 0x07 again in status reg to
> check for updated
> > lec later-on. In my view seems logical to add it also.
>
> It's done in cc770_status_interrupt():
>
> status = cc770_read_reg(priv, status);
> /* Reset the status register including RXOK and TXOK */
> cc770_write_reg(priv, status, STAT_LEC_MASK);
>
Oops.. Sorry it seems I missed it.
Many Thanks for your help and comments.
Regards,
Bhupesh
^ permalink raw reply
* [PATCH] pch_gbe: Fix the issue which a driver locks when rx offload is set by ethtool
From: Toshiharu Okada @ 2011-02-08 11:33 UTC (permalink / raw)
To: ML netdev, David S. Miller
Cc: Tomoya Morinaga, Wang, Qi, Wang, Yong Y, Andrew, Intel OTC,
Ewe, Kok Howg, LKML
This driver will be in a deadlock, When the rx offload is set by ethtool.
So, The pch_gbe_reinit_locked function was modified.
Signed-off-by: Toshiharu Okada <toshiharu-linux@dsn.okisemi.com>
---
drivers/net/pch_gbe/pch_gbe_main.c | 10 ++++------
1 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/drivers/net/pch_gbe/pch_gbe_main.c b/drivers/net/pch_gbe/pch_gbe_main.c
index 3248313..0e6922f 100644
--- a/drivers/net/pch_gbe/pch_gbe_main.c
+++ b/drivers/net/pch_gbe/pch_gbe_main.c
@@ -520,7 +520,9 @@ static void pch_gbe_reset_task(struct work_struct *work)
struct pch_gbe_adapter *adapter;
adapter = container_of(work, struct pch_gbe_adapter, reset_task);
+ rtnl_lock();
pch_gbe_reinit_locked(adapter);
+ rtnl_unlock();
}
/**
@@ -531,12 +533,8 @@ void pch_gbe_reinit_locked(struct pch_gbe_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
- rtnl_lock();
- if (netif_running(netdev)) {
- pch_gbe_down(adapter);
- pch_gbe_up(adapter);
- }
- rtnl_unlock();
+ pch_gbe_down(adapter);
+ pch_gbe_up(adapter);
}
/**
--
1.6.2.5
^ permalink raw reply related
* Re: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support for Bosch C_CAN controller
From: Wolfgang Grandegger @ 2011-02-08 11:28 UTC (permalink / raw)
To: Bhupesh SHARMA
Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org,
netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Marc Kleine-Budde
In-Reply-To: <D5ECB3C7A6F99444980976A8C6D896384DEE365EE8-8vAmw3ZAcdzhJTuQ9jeba9BPR1lH4CV8@public.gmane.org>
On 02/08/2011 11:45 AM, Bhupesh SHARMA wrote:
>> -----Original Message-----
>> From: Wolfgang Grandegger [mailto:wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org]
>> Sent: Tuesday, February 08, 2011 4:01 PM
>> To: Bhupesh SHARMA
>> Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org; netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; Marc
>> Kleine-Budde
>> Subject: Re: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support for
>> Bosch C_CAN controller
>>
>> On 02/08/2011 10:04 AM, Bhupesh SHARMA wrote:
>>> Hi Wolfgang,
>>>
>>>>> + stats->rx_errors++;
>>>>> + cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
>>>>> + cf->data[2] |= CAN_ERR_PROT_UNSPEC;
>>>>> +
>>>>> + switch (lec_type) {
>>>>> + case LEC_STUFF_ERROR:
>>>>> + netdev_dbg(dev, "stuff error\n");
>>>>> + cf->data[2] |= CAN_ERR_PROT_STUFF;
>>>>> + break;
>>>>> +
>>>>> + case LEC_FORM_ERROR:
>>>>> + netdev_dbg(dev, "form error\n");
>>>>> + cf->data[2] |= CAN_ERR_PROT_FORM;
>>>>> + break;
>>>>> +
>>>>> + case LEC_ACK_ERROR:
>>>>> + netdev_dbg(dev, "ack error\n");
>>>>> + cf->data[2] |= (CAN_ERR_PROT_LOC_ACK |
>>>>> + CAN_ERR_PROT_LOC_ACK_DEL);
>>>>> + break;
>>>>> +
>>>>> + case LEC_BIT1_ERROR:
>>>>> + netdev_dbg(dev, "bit1 error\n");
>>>>> + cf->data[2] |= CAN_ERR_PROT_BIT1;
>>>>> + break;
>>>>> +
>>>>> + case LEC_BIT0_ERROR:
>>>>> + netdev_dbg(dev, "bit0 error\n");
>>>>> + cf->data[2] |= CAN_ERR_PROT_BIT0;
>>>>> + break;
>>>>> +
>>>>> + case LEC_CRC_ERROR:
>>>>> + netdev_dbg(dev, "CRC error\n");
>>>>> + cf->data[2] |= (CAN_ERR_PROT_LOC_CRC_SEQ |
>>>>> + CAN_ERR_PROT_LOC_CRC_DEL);
>>>>> + break;
>>>>> + }
>>>
>>> >From the C_CAN manual:
>>>>
>>>> "The LEC field holds a code which indicates the type of the last
>> error
>>>> to occur on the CAN bus. This field will be cleared to '0' when a
>>>> message has been transferred (reception or transmission) without
>> error.
>>>> The unused code '7' may be written by the CPU to check for updates."
>>>
>>>> Not sure if it's necessary to reset the lec at init and after an
>> error
>>>> to 0x7 and check it. More below...
>>>
>>> I worked on your suggestion and instead found that the follow algo
>> must be used
>>> for reading updated `lec` values:
>>> a. Init lec by 0x07 at start.
>>> b. In function `c_can_err` check if lec is 0x7 (no bus error since
>> this value was
>>> written by CPU on status register) or 0x0 (no error). If so, return
>> without
>>> sending an error frame on stack. Else, check for the lec error type
>> and
>>> submit error frame on stack accordingly.
>>> c. In case a lec error is found and served in `c_can_err` routine,
>> write
>>> lec value to 0x07 again in status reg to check for updated lec later.
>>>
>>> This is my understanding after reading the specs time and again and
>>> implementing/testing the logic.
>>>
>>> Do you think this is fine or do you have any better approach?
>>
>> That's what I remember from the CC770 driver. Search for lec in:
>>
>> http://svn.berlios.de/wsvn/socketcan/trunk/kernel/2.6/drivers/net/can/c
>> c770/cc770.c
>>
>
> Seems similar. But step (c) mentioned above seems missing from cc770.c driver,
> i.e. "In case a lec error is found and served (by means of sending an error
> frame on the bus) write lec value to 0x07 again in status reg to check for updated
> lec later-on. In my view seems logical to add it also.
It's done in cc770_status_interrupt():
status = cc770_read_reg(priv, status);
/* Reset the status register including RXOK and TXOK */
cc770_write_reg(priv, status, STAT_LEC_MASK);
Wolfgang.
^ permalink raw reply
* RE: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support for Bosch C_CAN controller
From: Bhupesh SHARMA @ 2011-02-08 10:45 UTC (permalink / raw)
To: Wolfgang Grandegger
Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org,
netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Marc Kleine-Budde
In-Reply-To: <4D511B4E.1090001-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
> -----Original Message-----
> From: Wolfgang Grandegger [mailto:wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org]
> Sent: Tuesday, February 08, 2011 4:01 PM
> To: Bhupesh SHARMA
> Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org; netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org; Marc
> Kleine-Budde
> Subject: Re: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support for
> Bosch C_CAN controller
>
> On 02/08/2011 10:04 AM, Bhupesh SHARMA wrote:
> > Hi Wolfgang,
> >
> >>> + stats->rx_errors++;
> >>> + cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
> >>> + cf->data[2] |= CAN_ERR_PROT_UNSPEC;
> >>> +
> >>> + switch (lec_type) {
> >>> + case LEC_STUFF_ERROR:
> >>> + netdev_dbg(dev, "stuff error\n");
> >>> + cf->data[2] |= CAN_ERR_PROT_STUFF;
> >>> + break;
> >>> +
> >>> + case LEC_FORM_ERROR:
> >>> + netdev_dbg(dev, "form error\n");
> >>> + cf->data[2] |= CAN_ERR_PROT_FORM;
> >>> + break;
> >>> +
> >>> + case LEC_ACK_ERROR:
> >>> + netdev_dbg(dev, "ack error\n");
> >>> + cf->data[2] |= (CAN_ERR_PROT_LOC_ACK |
> >>> + CAN_ERR_PROT_LOC_ACK_DEL);
> >>> + break;
> >>> +
> >>> + case LEC_BIT1_ERROR:
> >>> + netdev_dbg(dev, "bit1 error\n");
> >>> + cf->data[2] |= CAN_ERR_PROT_BIT1;
> >>> + break;
> >>> +
> >>> + case LEC_BIT0_ERROR:
> >>> + netdev_dbg(dev, "bit0 error\n");
> >>> + cf->data[2] |= CAN_ERR_PROT_BIT0;
> >>> + break;
> >>> +
> >>> + case LEC_CRC_ERROR:
> >>> + netdev_dbg(dev, "CRC error\n");
> >>> + cf->data[2] |= (CAN_ERR_PROT_LOC_CRC_SEQ |
> >>> + CAN_ERR_PROT_LOC_CRC_DEL);
> >>> + break;
> >>> + }
> >
> >>From the C_CAN manual:
> >>
> >> "The LEC field holds a code which indicates the type of the last
> error
> >> to occur on the CAN bus. This field will be cleared to '0' when a
> >> message has been transferred (reception or transmission) without
> error.
> >> The unused code '7' may be written by the CPU to check for updates."
> >
> >> Not sure if it's necessary to reset the lec at init and after an
> error
> >> to 0x7 and check it. More below...
> >
> > I worked on your suggestion and instead found that the follow algo
> must be used
> > for reading updated `lec` values:
> > a. Init lec by 0x07 at start.
> > b. In function `c_can_err` check if lec is 0x7 (no bus error since
> this value was
> > written by CPU on status register) or 0x0 (no error). If so, return
> without
> > sending an error frame on stack. Else, check for the lec error type
> and
> > submit error frame on stack accordingly.
> > c. In case a lec error is found and served in `c_can_err` routine,
> write
> > lec value to 0x07 again in status reg to check for updated lec later.
> >
> > This is my understanding after reading the specs time and again and
> > implementing/testing the logic.
> >
> > Do you think this is fine or do you have any better approach?
>
> That's what I remember from the CC770 driver. Search for lec in:
>
> http://svn.berlios.de/wsvn/socketcan/trunk/kernel/2.6/drivers/net/can/c
> c770/cc770.c
>
Seems similar. But step (c) mentioned above seems missing from cc770.c driver,
i.e. "In case a lec error is found and served (by means of sending an error
frame on the bus) write lec value to 0x07 again in status reg to check for updated
lec later-on. In my view seems logical to add it also.
Regards,
Bhupesh
^ permalink raw reply
* Re: [PATCH net-next-2.6 v5 1/1] can: c_can: Added support for Bosch C_CAN controller
From: Wolfgang Grandegger @ 2011-02-08 10:30 UTC (permalink / raw)
To: Bhupesh SHARMA
Cc: socketcan-core-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org,
netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Marc Kleine-Budde
In-Reply-To: <D5ECB3C7A6F99444980976A8C6D896384DEE2BE326-8vAmw3ZAcdzhJTuQ9jeba9BPR1lH4CV8@public.gmane.org>
On 02/08/2011 10:04 AM, Bhupesh SHARMA wrote:
> Hi Wolfgang,
>
>>> + stats->rx_errors++;
>>> + cf->can_id |= CAN_ERR_PROT | CAN_ERR_BUSERROR;
>>> + cf->data[2] |= CAN_ERR_PROT_UNSPEC;
>>> +
>>> + switch (lec_type) {
>>> + case LEC_STUFF_ERROR:
>>> + netdev_dbg(dev, "stuff error\n");
>>> + cf->data[2] |= CAN_ERR_PROT_STUFF;
>>> + break;
>>> +
>>> + case LEC_FORM_ERROR:
>>> + netdev_dbg(dev, "form error\n");
>>> + cf->data[2] |= CAN_ERR_PROT_FORM;
>>> + break;
>>> +
>>> + case LEC_ACK_ERROR:
>>> + netdev_dbg(dev, "ack error\n");
>>> + cf->data[2] |= (CAN_ERR_PROT_LOC_ACK |
>>> + CAN_ERR_PROT_LOC_ACK_DEL);
>>> + break;
>>> +
>>> + case LEC_BIT1_ERROR:
>>> + netdev_dbg(dev, "bit1 error\n");
>>> + cf->data[2] |= CAN_ERR_PROT_BIT1;
>>> + break;
>>> +
>>> + case LEC_BIT0_ERROR:
>>> + netdev_dbg(dev, "bit0 error\n");
>>> + cf->data[2] |= CAN_ERR_PROT_BIT0;
>>> + break;
>>> +
>>> + case LEC_CRC_ERROR:
>>> + netdev_dbg(dev, "CRC error\n");
>>> + cf->data[2] |= (CAN_ERR_PROT_LOC_CRC_SEQ |
>>> + CAN_ERR_PROT_LOC_CRC_DEL);
>>> + break;
>>> + }
>
>>From the C_CAN manual:
>>
>> "The LEC field holds a code which indicates the type of the last error
>> to occur on the CAN bus. This field will be cleared to '0' when a
>> message has been transferred (reception or transmission) without error.
>> The unused code '7' may be written by the CPU to check for updates."
>
>> Not sure if it's necessary to reset the lec at init and after an error
>> to 0x7 and check it. More below...
>
> I worked on your suggestion and instead found that the follow algo must be used
> for reading updated `lec` values:
> a. Init lec by 0x07 at start.
> b. In function `c_can_err` check if lec is 0x7 (no bus error since this value was
> written by CPU on status register) or 0x0 (no error). If so, return without
> sending an error frame on stack. Else, check for the lec error type and
> submit error frame on stack accordingly.
> c. In case a lec error is found and served in `c_can_err` routine, write
> lec value to 0x07 again in status reg to check for updated lec later.
>
> This is my understanding after reading the specs time and again and
> implementing/testing the logic.
>
> Do you think this is fine or do you have any better approach?
That's what I remember from the CC770 driver. Search for lec in:
http://svn.berlios.de/wsvn/socketcan/trunk/kernel/2.6/drivers/net/can/cc770/cc770.c
Wolfgang.
^ permalink raw reply
* [PATCH v2] xen network backend driver
From: Ian Campbell @ 2011-02-08 10:23 UTC (permalink / raw)
To: netdev@vger.kernel.org, xen-devel
Cc: Jeremy Fitzhardinge, Konrad Rzeszutek Wilk, Ben Hutchings,
Herbert Xu
Hi again,
The following patch is the second iteration of the Xen network backend
driver for upstream Linux.
This driver ("netback") is the host side counterpart to the frontend
driver in drivers/net/xen-netfront.c. The PV protocol is also
implemented by frontend drivers in other OSes too, such as the BSDs and
even Windows.
This driver has a long history as an out of tree driver but I am
submitting it here as a single monolithic patch to aid review. Once it
has been reviewed and is considered suitable for merging can we perhaps
consider merging the equivalent git branch which maintains much of
history?
Changes since the last posting, many due to Ben Hutching's review,
include:
* Improved Kconfig description for XEN_NETDEV_BACKEND and
XEN_NETDEV_FRONTEND.
* Avoid the core networking namespaces (skb_*, netif_*, net_*).
This led to a major refactoring since the current namespace use
was something of a mess. Now the code tries to consistently use
xenvif* for the device driver related stuff (interface.c) and
xen_netbk* for the backend worker pool related stuff
(netback.c). This cleanup extended to the
xen/interface/io/netif.h header which required changes to
netfront too.
* Dropped the tasklet mode for the backend worker leaving only the
kthread mode. I will revisit the suggestion to use NAPI on the
driver side in the future, I think it's somewhat orthogonal to
the use of kthread here, but it seems likely to be a worthwhile
improvement either way.
* Dropped netbk_copy_skb. Ben requested this function be made
generic and moved to the networking core but it turns out it was
trivial to remove netback's reliance on this functionality, and
avoid a bunch of unnecessary copying in the process. The
function's semantics were a bit odd in any case so I couldn't
imagine many other users.
* Handle incoming GSO SKBs which are not CHECKSUM_PARTIAL
correctly. Changed from previous behaviour (dropping the skb) to
doing a fixup after discussion of equivalent frontend patch
which became e0ce4af920eb028f38bfd680b1d733f4c7a0b7cf.
* Other improvements suggested by Ben (e.g. dropping pointless
filename references from top of file comments, not including
version.h, correct return values from ethtool hooks, dropped
queue_length module parameter, dropped unused debug interrupt,
etc)
Changes made for the initial upstream post of the driver vs. the out of
tree xen.git pvops version include:
* The driver has been put through the checkpatch.pl wringer plus
several manual cleanup passes.
* Moved from drivers/xen/netback to drivers/net/xen-netback.
* Most significantly the guest transmit path (i.e. what looks like
receive to netback) has been significantly reworked to remove
the dependency on the out of tree PageForeign page flag (a core
kernel patch which enables a per page destructor callback on the
final put_page). This page flag is needed in order to implement
a grant map based transmit path (where guest pages are mapped
directly into SKB frags). Instead this version of netback uses
grant copy operations into regular memory belonging to the
backend domain. Reinstating the grant map functionality is
something which I would like to revisit in the future.
The series is also available in git at
git://xenbits.xen.org/people/ianc/linux-2.6.git upstream/dom0/backend/netback
(based on mainline 329620a878cf89184b28500d37fa33cc870a3357)
The upstream/dom0/backend/netback-base branch contains the history which
is imported from the xen.git tree. This is followed by
upstream/dom0/backend/netback-cleanup which is the checkpatch and other
cleanups and finally upstream/dom0/backend/netback has the upstreaming
specific changes.
The complete patch's diffstat looks like:
drivers/net/Kconfig | 38 +-
drivers/net/Makefile | 1 +
drivers/net/xen-netback/Makefile | 3 +
drivers/net/xen-netback/common.h | 147 ++++
drivers/net/xen-netback/interface.c | 550 ++++++++++++
drivers/net/xen-netback/netback.c | 1618 +++++++++++++++++++++++++++++++++++
drivers/net/xen-netback/xenbus.c | 490 +++++++++++
drivers/net/xen-netfront.c | 20 +-
include/xen/interface/io/netif.h | 80 +-
9 files changed, 2893 insertions(+), 54 deletions(-)
To give an idea how much has changed versus the xen.git version the
diffstat between upstream/dom0/backend/netback-base and
upstream/dom0/backend/netback is:
drivers/net/Kconfig | 38 +-
drivers/net/Makefile | 1 +
drivers/{xen/netback => net/xen-netback}/Makefile | 0
drivers/net/xen-netback/common.h | 147 ++
drivers/net/xen-netback/interface.c | 550 ++++++
drivers/net/xen-netback/netback.c | 1618 ++++++++++++++++++
drivers/{xen/netback => net/xen-netback}/xenbus.c | 155 +-
drivers/net/xen-netfront.c | 20 +-
drivers/xen/Kconfig | 7 -
drivers/xen/Makefile | 1 -
drivers/xen/netback/common.h | 326 ----
drivers/xen/netback/interface.c | 471 -----
drivers/xen/netback/netback.c | 1902 ---------------------
include/xen/interface/io/netif.h | 80 +-
14 files changed, 2468 insertions(+), 2848 deletions(-)
Would it be useful (mainly to xen-devel I guess) to post these cleanup
patches as a series?
Also, I can separate out the netfront bits (Kconfig help update and
netif.h changes) if that would be preferred.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index cbf0635..1c77e18 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2963,12 +2963,38 @@ config XEN_NETDEV_FRONTEND
select XEN_XENBUS_FRONTEND
default y
help
- The network device frontend driver allows the kernel to
- access network devices exported exported by a virtual
- machine containing a physical network device driver. The
- frontend driver is intended for unprivileged guest domains;
- if you are compiling a kernel for a Xen guest, you almost
- certainly want to enable this.
+ This driver provides support for Xen paravirtual network
+ devices exported by a Xen network driver domain (often
+ domain 0).
+
+ The corresponding Linux backend driver is enabled by the
+ CONFIG_XEN_NETDEV_BACKEND option.
+
+ If you are compiling a kernel for use as Xen guest, you
+ should say Y here. To compile this driver as a module, chose
+ M here: the module will be called xen-netfront.
+
+config XEN_NETDEV_BACKEND
+ tristate "Xen backend network device"
+ depends on XEN_BACKEND
+ help
+ This driver allows the kernel to act as a Xen network driver
+ domain which exports paravirtual network devices to other
+ Xen domains. These devices can be accessed by any operating
+ system that implements a compatible front end.
+
+ The corresponding Linux frontend driver is enabled by the
+ CONFIG_XEN_NETDEV_FRONTEND configuration option.
+
+ The backend driver presents a standard network device
+ endpoint for each paravirtual network device to the driver
+ domain network stack. These can then be bridged or routed
+ etc in order to provide full network connectivity.
+
+ If you are compiling a kernel to run in a Xen network driver
+ domain (often this is domain 0) you should say Y here. To
+ compile this driver as a module, chose M here: the module
+ will be called xen-netback.
config ISERIES_VETH
tristate "iSeries Virtual Ethernet driver support"
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index b90738d..145dfd7 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -171,6 +171,7 @@ obj-$(CONFIG_SLIP) += slip.o
obj-$(CONFIG_SLHC) += slhc.o
obj-$(CONFIG_XEN_NETDEV_FRONTEND) += xen-netfront.o
+obj-$(CONFIG_XEN_NETDEV_BACKEND) += xen-netback/
obj-$(CONFIG_DUMMY) += dummy.o
obj-$(CONFIG_IFB) += ifb.o
diff --git a/drivers/net/xen-netback/Makefile b/drivers/net/xen-netback/Makefile
new file mode 100644
index 0000000..e346e81
--- /dev/null
+++ b/drivers/net/xen-netback/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_XEN_NETDEV_BACKEND) := xen-netback.o
+
+xen-netback-y := netback.o xenbus.o interface.o
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
new file mode 100644
index 0000000..03196ab
--- /dev/null
+++ b/drivers/net/xen-netback/common.h
@@ -0,0 +1,147 @@
+/*
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#ifndef __XEN_NETBACK__COMMON_H__
+#define __XEN_NETBACK__COMMON_H__
+
+#define pr_fmt(fmt) KBUILD_MODNAME ":%s: " fmt, __func__
+
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+#include <linux/ip.h>
+#include <linux/in.h>
+#include <linux/io.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/wait.h>
+#include <linux/sched.h>
+
+#include <xen/interface/io/netif.h>
+#include <asm/pgalloc.h>
+#include <xen/interface/grant_table.h>
+#include <xen/grant_table.h>
+#include <xen/xenbus.h>
+
+struct xen_netbk;
+
+struct xenvif {
+ /* Unique identifier for this interface. */
+ domid_t domid;
+ unsigned int handle;
+
+ /* */
+ struct xen_netbk *netbk;
+
+ u8 fe_dev_addr[6];
+
+ /* Physical parameters of the comms window. */
+ grant_handle_t tx_shmem_handle;
+ grant_ref_t tx_shmem_ref;
+ grant_handle_t rx_shmem_handle;
+ grant_ref_t rx_shmem_ref;
+ unsigned int irq;
+
+ /* The shared rings and indexes. */
+ struct xen_netif_tx_back_ring tx;
+ struct xen_netif_rx_back_ring rx;
+ struct vm_struct *tx_comms_area;
+ struct vm_struct *rx_comms_area;
+
+ /* Flags that must not be set in dev->features */
+ int features_disabled;
+
+ /* Frontend feature information. */
+ u8 can_sg:1;
+ u8 gso:1;
+ u8 gso_prefix:1;
+ u8 csum:1;
+
+ /* Internal feature information. */
+ u8 can_queue:1; /* can queue packets for receiver? */
+
+ /* Allow xenvif_start_xmit() to peek ahead in the rx request
+ * ring. This is a prediction of what rx_req_cons will be once
+ * all queued skbs are put on the ring. */
+ RING_IDX rx_req_cons_peek;
+
+ /* Transmit shaping: allow 'credit_bytes' every 'credit_usec'. */
+ unsigned long credit_bytes;
+ unsigned long credit_usec;
+ unsigned long remaining_credit;
+ struct timer_list credit_timeout;
+
+ /* Statistics */
+ int rx_gso_checksum_fixup;
+
+ /* Miscellaneous private stuff. */
+ struct list_head list; /* scheduling list */
+ atomic_t refcnt;
+ struct net_device *dev;
+ struct net_device_stats stats;
+
+ unsigned int carrier;
+
+ wait_queue_head_t waiting_to_free;
+};
+
+#define XEN_NETIF_TX_RING_SIZE __RING_SIZE((struct xen_netif_tx_sring *)0, PAGE_SIZE)
+#define XEN_NETIF_RX_RING_SIZE __RING_SIZE((struct xen_netif_rx_sring *)0, PAGE_SIZE)
+
+struct xenvif *xenvif_alloc(struct device *parent,
+ domid_t domid,
+ unsigned int handle);
+
+int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
+ unsigned long rx_ring_ref, unsigned int evtchn);
+void xenvif_disconnect(struct xenvif *vif);
+
+void xenvif_get(struct xenvif *vif);
+void xenvif_put(struct xenvif *vif);
+
+int xenvif_xenbus_init(void);
+
+int xenvif_schedulable(struct xenvif *vif);
+
+void xenvif_schedule_work(struct xenvif *vif);
+
+int xenvif_queue_full(struct xenvif *vif);
+
+/* (De)Register a xenvif with the netback backend. */
+void xen_netbk_add_xenvif(struct xenvif *vif);
+void xen_netbk_remove_xenvif(struct xenvif *vif);
+
+/* */
+void xen_netbk_schedule_xenvif(struct xenvif *vif);
+void xen_netbk_deschedule_xenfif(struct xenvif *vif);
+
+/* */
+unsigned int xen_netbk_count_skb_slots(struct xenvif *vif, struct sk_buff *skb);
+
+/* */
+void xen_netbk_queue_tx_skb(struct xenvif *vif, struct sk_buff *skb);
+
+#endif /* __XEN_NETBACK__COMMON_H__ */
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
new file mode 100644
index 0000000..98a992d
--- /dev/null
+++ b/drivers/net/xen-netback/interface.c
@@ -0,0 +1,550 @@
+/*
+ * Network-device interface management.
+ *
+ * Copyright (c) 2004-2005, Keir Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include "common.h"
+
+#include <linux/ethtool.h>
+#include <linux/rtnetlink.h>
+
+#include <xen/events.h>
+#include <asm/xen/hypercall.h>
+
+#define XENVIF_QUEUE_LENGTH 32
+
+void xenvif_get(struct xenvif *vif)
+{
+ atomic_inc(&vif->refcnt);
+}
+
+void xenvif_put(struct xenvif *vif)
+{
+ if (atomic_dec_and_test(&vif->refcnt))
+ wake_up(&vif->waiting_to_free);
+}
+
+static int xenvif_max_required_rx_slots(struct xenvif *vif)
+{
+ int max = DIV_ROUND_UP(vif->dev->mtu, PAGE_SIZE);
+
+ if (vif->can_sg || vif->gso || vif->gso_prefix)
+ max += MAX_SKB_FRAGS + 1; /* extra_info + frags */
+
+ return max;
+}
+
+int xenvif_queue_full(struct xenvif *vif)
+{
+ RING_IDX peek = vif->rx_req_cons_peek;
+ RING_IDX needed = xenvif_max_required_rx_slots(vif);
+
+ return ((vif->rx.sring->req_prod - peek) < needed) ||
+ ((vif->rx.rsp_prod_pvt + XEN_NETIF_RX_RING_SIZE - peek) < needed);
+}
+
+/*
+ * Implement our own carrier flag: the network stack's version causes delays
+ * when the carrier is re-enabled (in particular, dev_activate() may not
+ * immediately be called, which can cause packet loss; also the etherbridge
+ * can be rather lazy in activating its port).
+ */
+static void xenvif_carrier_on(struct xenvif *vif)
+{
+ vif->carrier = 1;
+}
+static void xenvif_carrier_off(struct xenvif *vif)
+{
+ vif->carrier = 0;
+}
+static int xenvif_carrier_ok(struct xenvif *vif)
+{
+ return vif->carrier;
+}
+
+int xenvif_schedulable(struct xenvif *vif)
+{
+ return netif_running(vif->dev) && xenvif_carrier_ok(vif);
+}
+
+static irqreturn_t xenvif_interrupt(int irq, void *dev_id)
+{
+ struct xenvif *vif = dev_id;
+
+ if (vif->netbk == NULL)
+ return IRQ_NONE;
+
+ xen_netbk_schedule_xenvif(vif);
+
+ if (xenvif_schedulable(vif) && !xenvif_queue_full(vif))
+ netif_wake_queue(vif->dev);
+
+ return IRQ_HANDLED;
+}
+
+static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct xenvif *vif = netdev_priv(dev);
+
+ BUG_ON(skb->dev != dev);
+
+ if (vif->netbk == NULL)
+ goto drop;
+
+ /* Drop the packet if the target domain has no receive buffers. */
+ if (unlikely(!xenvif_schedulable(vif) || xenvif_queue_full(vif)))
+ goto drop;
+
+ /* Reserve ring slots for the worst-case number of fragments. */
+ vif->rx_req_cons_peek += xen_netbk_count_skb_slots(vif, skb);
+ xenvif_get(vif);
+
+ if (vif->can_queue && xenvif_queue_full(vif)) {
+ vif->rx.sring->req_event = vif->rx_req_cons_peek +
+ xenvif_max_required_rx_slots(vif);
+ mb(); /* request notification /then/ check & stop the queue */
+ if (xenvif_queue_full(vif))
+ netif_stop_queue(dev);
+ }
+
+ xen_netbk_queue_tx_skb(vif, skb);
+
+ return 0;
+
+ drop:
+ vif->stats.tx_dropped++;
+ dev_kfree_skb(skb);
+ return 0;
+}
+
+static struct net_device_stats *xenvif_get_stats(struct net_device *dev)
+{
+ struct xenvif *vif = netdev_priv(dev);
+ return &vif->stats;
+}
+
+void xenvif_schedule_work(struct xenvif *vif)
+{
+ int more_to_do;
+
+ RING_FINAL_CHECK_FOR_REQUESTS(&vif->tx, more_to_do);
+
+ if (more_to_do)
+ xen_netbk_schedule_xenvif(vif);
+}
+
+
+static void xenvif_up(struct xenvif *vif)
+{
+ xen_netbk_add_xenvif(vif);
+ enable_irq(vif->irq);
+ xenvif_schedule_work(vif);
+}
+
+static void xenvif_down(struct xenvif *vif)
+{
+ disable_irq(vif->irq);
+ xen_netbk_deschedule_xenfif(vif);
+ xen_netbk_remove_xenvif(vif);
+}
+
+static int xenvif_open(struct net_device *dev)
+{
+ struct xenvif *vif = netdev_priv(dev);
+ if (xenvif_carrier_ok(vif)) {
+ xenvif_up(vif);
+ netif_start_queue(dev);
+ }
+ return 0;
+}
+
+static int xenvif_close(struct net_device *dev)
+{
+ struct xenvif *vif = netdev_priv(dev);
+ if (xenvif_carrier_ok(vif))
+ xenvif_down(vif);
+ netif_stop_queue(dev);
+ return 0;
+}
+
+static int xenvif_change_mtu(struct net_device *dev, int mtu)
+{
+ struct xenvif *vif = netdev_priv(dev);
+ int max = vif->can_sg ? 65535 - ETH_HLEN : ETH_DATA_LEN;
+
+ if (mtu > max)
+ return -EINVAL;
+ dev->mtu = mtu;
+ return 0;
+}
+
+static void xenvif_set_features(struct xenvif *vif)
+{
+ struct net_device *dev = vif->dev;
+ int features = dev->features;
+
+ if (vif->can_sg)
+ features |= NETIF_F_SG;
+ if (vif->gso || vif->gso_prefix)
+ features |= NETIF_F_TSO;
+ if (vif->csum)
+ features |= NETIF_F_IP_CSUM;
+
+ features &= ~(vif->features_disabled);
+
+ if (!(features & NETIF_F_SG) && dev->mtu > ETH_DATA_LEN)
+ dev->mtu = ETH_DATA_LEN;
+
+ dev->features = features;
+}
+
+static int xenvif_set_tx_csum(struct net_device *dev, u32 data)
+{
+ struct xenvif *vif = netdev_priv(dev);
+ if (data) {
+ if (!vif->csum)
+ return -EOPNOTSUPP;
+ vif->features_disabled &= ~NETIF_F_IP_CSUM;
+ } else {
+ vif->features_disabled |= NETIF_F_IP_CSUM;
+ }
+
+ xenvif_set_features(vif);
+ return 0;
+}
+
+static int xenvif_set_sg(struct net_device *dev, u32 data)
+{
+ struct xenvif *vif = netdev_priv(dev);
+ if (data) {
+ if (!vif->can_sg)
+ return -EOPNOTSUPP;
+ vif->features_disabled &= ~NETIF_F_SG;
+ } else {
+ vif->features_disabled |= NETIF_F_SG;
+ }
+
+ xenvif_set_features(vif);
+ return 0;
+}
+
+static int xenvif_set_tso(struct net_device *dev, u32 data)
+{
+ struct xenvif *vif = netdev_priv(dev);
+ if (data) {
+ if (!vif->gso && !vif->gso_prefix)
+ return -EOPNOTSUPP;
+ vif->features_disabled &= ~NETIF_F_TSO;
+ } else {
+ vif->features_disabled |= NETIF_F_TSO;
+ }
+
+ xenvif_set_features(vif);
+ return 0;
+}
+
+static const struct xenvif_stat {
+ char name[ETH_GSTRING_LEN];
+ u16 offset;
+} xenvif_stats[] = {
+ {
+ "rx_gso_checksum_fixup",
+ offsetof(struct xenvif, rx_gso_checksum_fixup)
+ },
+};
+
+static int xenvif_get_sset_count(struct net_device *dev, int string_set)
+{
+ switch (string_set) {
+ case ETH_SS_STATS:
+ return ARRAY_SIZE(xenvif_stats);
+ default:
+ return -EINVAL;
+ }
+}
+
+static void xenvif_get_ethtool_stats(struct net_device *dev,
+ struct ethtool_stats *stats, u64 * data)
+{
+ void *vif = netdev_priv(dev);
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(xenvif_stats); i++)
+ data[i] = *(int *)(vif + xenvif_stats[i].offset);
+}
+
+static void xenvif_get_strings(struct net_device *dev, u32 stringset, u8 * data)
+{
+ int i;
+
+ switch (stringset) {
+ case ETH_SS_STATS:
+ for (i = 0; i < ARRAY_SIZE(xenvif_stats); i++)
+ memcpy(data + i * ETH_GSTRING_LEN,
+ xenvif_stats[i].name, ETH_GSTRING_LEN);
+ break;
+ }
+}
+
+static struct ethtool_ops xenvif_ethtool_ops = {
+ .get_tx_csum = ethtool_op_get_tx_csum,
+ .set_tx_csum = xenvif_set_tx_csum,
+ .get_sg = ethtool_op_get_sg,
+ .set_sg = xenvif_set_sg,
+ .get_tso = ethtool_op_get_tso,
+ .set_tso = xenvif_set_tso,
+ .get_link = ethtool_op_get_link,
+
+ .get_sset_count = xenvif_get_sset_count,
+ .get_ethtool_stats = xenvif_get_ethtool_stats,
+ .get_strings = xenvif_get_strings,
+};
+
+static struct net_device_ops xenvif_netdev_ops = {
+ .ndo_start_xmit = xenvif_start_xmit,
+ .ndo_get_stats = xenvif_get_stats,
+ .ndo_open = xenvif_open,
+ .ndo_stop = xenvif_close,
+ .ndo_change_mtu = xenvif_change_mtu,
+};
+
+struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
+ unsigned int handle)
+{
+ int err = 0;
+ struct net_device *dev;
+ struct xenvif *vif;
+ char name[IFNAMSIZ] = {};
+
+ snprintf(name, IFNAMSIZ - 1, "vif%u.%u", domid, handle);
+ dev = alloc_netdev(sizeof(struct xenvif), name, ether_setup);
+ if (dev == NULL) {
+ pr_debug("Could not allocate netdev\n");
+ return ERR_PTR(-ENOMEM);
+ }
+
+ SET_NETDEV_DEV(dev, parent);
+
+ vif = netdev_priv(dev);
+ memset(vif, 0, sizeof(*vif));
+ vif->domid = domid;
+ vif->handle = handle;
+ vif->netbk = NULL;
+ vif->can_sg = 1;
+ vif->csum = 1;
+ atomic_set(&vif->refcnt, 1);
+ init_waitqueue_head(&vif->waiting_to_free);
+ vif->dev = dev;
+ INIT_LIST_HEAD(&vif->list);
+
+ xenvif_carrier_off(vif);
+
+ vif->credit_bytes = vif->remaining_credit = ~0UL;
+ vif->credit_usec = 0UL;
+ init_timer(&vif->credit_timeout);
+ /* Initialize 'expires' now: it's used to track the credit window. */
+ vif->credit_timeout.expires = jiffies;
+
+ dev->netdev_ops = &xenvif_netdev_ops;
+ xenvif_set_features(vif);
+ SET_ETHTOOL_OPS(dev, &xenvif_ethtool_ops);
+
+ dev->tx_queue_len = XENVIF_QUEUE_LENGTH;
+
+ /*
+ * Initialise a dummy MAC address. We choose the numerically
+ * largest non-broadcast address to prevent the address getting
+ * stolen by an Ethernet bridge for STP purposes.
+ * (FE:FF:FF:FF:FF:FF)
+ */
+ memset(dev->dev_addr, 0xFF, ETH_ALEN);
+ dev->dev_addr[0] &= ~0x01;
+
+ rtnl_lock();
+ err = register_netdevice(dev);
+ rtnl_unlock();
+ if (err) {
+ pr_debug("Could not register new net device %s: err=%d\n",
+ dev->name, err);
+ free_netdev(dev);
+ return ERR_PTR(err);
+ }
+
+ pr_debug("Successfully created xenvif\n");
+ return vif;
+}
+
+static int map_frontend_pages(struct xenvif *vif,
+ grant_ref_t tx_ring_ref,
+ grant_ref_t rx_ring_ref)
+{
+ struct gnttab_map_grant_ref op;
+
+ gnttab_set_map_op(&op, (unsigned long)vif->tx_comms_area->addr,
+ GNTMAP_host_map, tx_ring_ref, vif->domid);
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
+ BUG();
+
+ if (op.status) {
+ pr_debug("Gnttab failure mapping tx_ring_ref!\n");
+ return op.status;
+ }
+
+ vif->tx_shmem_ref = tx_ring_ref;
+ vif->tx_shmem_handle = op.handle;
+
+ gnttab_set_map_op(&op, (unsigned long)vif->rx_comms_area->addr,
+ GNTMAP_host_map, rx_ring_ref, vif->domid);
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
+ BUG();
+
+ if (op.status) {
+ struct gnttab_unmap_grant_ref unop;
+
+ gnttab_set_unmap_op(&unop,
+ (unsigned long)vif->tx_comms_area->addr,
+ GNTMAP_host_map, vif->tx_shmem_handle);
+ HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &unop, 1);
+ pr_debug("Gnttab failure mapping rx_ring_ref!\n");
+ return op.status;
+ }
+
+ vif->rx_shmem_ref = rx_ring_ref;
+ vif->rx_shmem_handle = op.handle;
+
+ return 0;
+}
+
+static void unmap_frontend_pages(struct xenvif *vif)
+{
+ struct gnttab_unmap_grant_ref op;
+
+ gnttab_set_unmap_op(&op, (unsigned long)vif->tx_comms_area->addr,
+ GNTMAP_host_map, vif->tx_shmem_handle);
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
+ BUG();
+
+ gnttab_set_unmap_op(&op, (unsigned long)vif->rx_comms_area->addr,
+ GNTMAP_host_map, vif->rx_shmem_handle);
+
+ if (HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1))
+ BUG();
+}
+
+int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
+ unsigned long rx_ring_ref, unsigned int evtchn)
+{
+ int err = -ENOMEM;
+ struct xen_netif_tx_sring *txs;
+ struct xen_netif_rx_sring *rxs;
+
+ /* Already connected through? */
+ if (vif->irq)
+ return 0;
+
+ xenvif_set_features(vif);
+
+ vif->tx_comms_area = alloc_vm_area(PAGE_SIZE);
+ if (vif->tx_comms_area == NULL)
+ return -ENOMEM;
+ vif->rx_comms_area = alloc_vm_area(PAGE_SIZE);
+ if (vif->rx_comms_area == NULL)
+ goto err_rx;
+
+ err = map_frontend_pages(vif, tx_ring_ref, rx_ring_ref);
+ if (err)
+ goto err_map;
+
+ err = bind_interdomain_evtchn_to_irqhandler(
+ vif->domid, evtchn, xenvif_interrupt, 0,
+ vif->dev->name, vif);
+ if (err < 0)
+ goto err_hypervisor;
+ vif->irq = err;
+ disable_irq(vif->irq);
+
+ txs = (struct xen_netif_tx_sring *)vif->tx_comms_area->addr;
+ BACK_RING_INIT(&vif->tx, txs, PAGE_SIZE);
+
+ rxs = (struct xen_netif_rx_sring *)
+ ((char *)vif->rx_comms_area->addr);
+ BACK_RING_INIT(&vif->rx, rxs, PAGE_SIZE);
+
+ vif->rx_req_cons_peek = 0;
+
+ xenvif_get(vif);
+
+ rtnl_lock();
+ xenvif_carrier_on(vif);
+ if (netif_running(vif->dev))
+ xenvif_up(vif);
+ rtnl_unlock();
+
+ return 0;
+err_hypervisor:
+ unmap_frontend_pages(vif);
+err_map:
+ free_vm_area(vif->rx_comms_area);
+err_rx:
+ free_vm_area(vif->tx_comms_area);
+ return err;
+}
+
+void xenvif_disconnect(struct xenvif *vif)
+{
+ if (xenvif_carrier_ok(vif)) {
+ rtnl_lock();
+ xenvif_carrier_off(vif);
+ netif_carrier_off(vif->dev); /* discard queued packets */
+ if (netif_running(vif->dev))
+ xenvif_down(vif);
+ rtnl_unlock();
+ xenvif_put(vif);
+ }
+
+ atomic_dec(&vif->refcnt);
+ wait_event(vif->waiting_to_free, atomic_read(&vif->refcnt) == 0);
+
+ del_timer_sync(&vif->credit_timeout);
+
+ if (vif->irq)
+ unbind_from_irqhandler(vif->irq, vif);
+
+ unregister_netdev(vif->dev);
+
+ if (vif->tx.sring) {
+ unmap_frontend_pages(vif);
+ free_vm_area(vif->tx_comms_area);
+ free_vm_area(vif->rx_comms_area);
+ }
+
+ free_netdev(vif->dev);
+}
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
new file mode 100644
index 0000000..fbddf3d
--- /dev/null
+++ b/drivers/net/xen-netback/netback.c
@@ -0,0 +1,1618 @@
+/*
+ * Back-end of the driver for virtual network devices. This portion of the
+ * driver exports a 'unified' network-device interface that can be accessed
+ * by any operating system that implements a compatible front end. A
+ * reference front-end implementation can be found in:
+ * drivers/net/xen-netfront.c
+ *
+ * Copyright (c) 2002-2005, K A Fraser
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include "common.h"
+
+#include <linux/kthread.h>
+#include <linux/if_vlan.h>
+#include <linux/udp.h>
+
+#include <net/tcp.h>
+
+#include <xen/events.h>
+#include <xen/interface/memory.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+struct pending_tx_info {
+ struct xen_netif_tx_request req;
+ struct xenvif *vif;
+};
+typedef unsigned int pending_ring_idx_t;
+
+struct netbk_rx_meta {
+ int id;
+ int size;
+ int gso_size;
+};
+
+#define MAX_PENDING_REQS 256
+
+#define MAX_BUFFER_OFFSET PAGE_SIZE
+
+/* extra field used in struct page */
+union page_ext {
+ struct {
+#if BITS_PER_LONG < 64
+#define IDX_WIDTH 8
+#define GROUP_WIDTH (BITS_PER_LONG - IDX_WIDTH)
+ unsigned int group:GROUP_WIDTH;
+ unsigned int idx:IDX_WIDTH;
+#else
+ unsigned int group, idx;
+#endif
+ } e;
+ void *mapping;
+};
+
+struct xen_netbk {
+ wait_queue_head_t wq;
+ struct task_struct *task;
+
+ struct sk_buff_head rx_queue;
+ struct sk_buff_head tx_queue;
+
+ struct timer_list net_timer;
+
+ struct page *mmap_pages[MAX_PENDING_REQS];
+
+ pending_ring_idx_t pending_prod;
+ pending_ring_idx_t pending_cons;
+ struct list_head net_schedule_list;
+
+ /* Protect the net_schedule_list in netif. */
+ spinlock_t net_schedule_list_lock;
+
+ atomic_t netfront_count;
+
+ struct pending_tx_info pending_tx_info[MAX_PENDING_REQS];
+ struct gnttab_copy tx_copy_ops[MAX_PENDING_REQS];
+
+ u16 pending_ring[MAX_PENDING_REQS];
+
+ /*
+ * Each head or fragment can be up to 4096 bytes. Given
+ * MAX_BUFFER_OFFSET of 4096 the worst case is that each
+ * head/fragment uses 2 copy operation.
+ */
+ struct gnttab_copy grant_copy_op[2*XEN_NETIF_RX_RING_SIZE];
+ unsigned char rx_notify[NR_IRQS];
+ u16 notify_list[XEN_NETIF_RX_RING_SIZE];
+ struct netbk_rx_meta meta[2*XEN_NETIF_RX_RING_SIZE];
+};
+
+static struct xen_netbk *xen_netbk;
+static int xen_netbk_group_nr;
+
+void xen_netbk_add_xenvif(struct xenvif *vif)
+{
+ int i;
+ int min_netfront_count;
+ int min_group = 0;
+ struct xen_netbk *netbk;
+
+ min_netfront_count = atomic_read(&xen_netbk[0].netfront_count);
+ for (i = 0; i < xen_netbk_group_nr; i++) {
+ int netfront_count = atomic_read(&xen_netbk[i].netfront_count);
+ if (netfront_count < min_netfront_count) {
+ min_group = i;
+ min_netfront_count = netfront_count;
+ }
+ }
+
+ netbk = &xen_netbk[min_group];
+
+ vif->netbk = netbk;
+ atomic_inc(&netbk->netfront_count);
+}
+
+void xen_netbk_remove_xenvif(struct xenvif *vif)
+{
+ struct xen_netbk *netbk = vif->netbk;
+ vif->netbk = NULL;
+ atomic_dec(&netbk->netfront_count);
+}
+
+static void xen_netbk_idx_release(struct xen_netbk *netbk, u16 pending_idx);
+static void make_tx_response(struct xenvif *vif,
+ struct xen_netif_tx_request *txp,
+ s8 st);
+static struct xen_netif_rx_response *make_rx_response(struct xenvif *vif,
+ u16 id,
+ s8 st,
+ u16 offset,
+ u16 size,
+ u16 flags);
+
+static inline unsigned long idx_to_pfn(struct xen_netbk *netbk,
+ unsigned int idx)
+{
+ return page_to_pfn(netbk->mmap_pages[idx]);
+}
+
+static inline unsigned long idx_to_kaddr(struct xen_netbk *netbk,
+ unsigned int idx)
+{
+ return (unsigned long)pfn_to_kaddr(idx_to_pfn(netbk, idx));
+}
+
+/* extra field used in struct page */
+static inline void set_page_ext(struct page *pg, struct xen_netbk *netbk,
+ unsigned int idx)
+{
+ unsigned int group = netbk - xen_netbk;
+ union page_ext ext = { .e = { .group = group + 1, .idx = idx } };
+
+ BUILD_BUG_ON(sizeof(ext) > sizeof(ext.mapping));
+ pg->mapping = ext.mapping;
+}
+
+static int get_page_ext(struct page *pg,
+ unsigned int *pgroup, unsigned int *pidx)
+{
+ union page_ext ext = { .mapping = pg->mapping };
+ struct xen_netbk *netbk;
+ unsigned int group, idx;
+
+ group = ext.e.group - 1;
+
+ if (group < 0 || group >= xen_netbk_group_nr)
+ return 0;
+
+ netbk = &xen_netbk[group];
+
+ idx = ext.e.idx;
+
+ if ((idx < 0) || (idx >= MAX_PENDING_REQS))
+ return 0;
+
+ if (netbk->mmap_pages[idx] != pg)
+ return 0;
+
+ *pgroup = group;
+ *pidx = idx;
+
+ return 1;
+}
+
+/*
+ * This is the amount of packet we copy rather than map, so that the
+ * guest can't fiddle with the contents of the headers while we do
+ * packet processing on them (netfilter, routing, etc).
+ */
+#define PKT_PROT_LEN (ETH_HLEN + \
+ VLAN_HLEN + \
+ sizeof(struct iphdr) + MAX_IPOPTLEN + \
+ sizeof(struct tcphdr) + MAX_TCP_OPTION_SPACE)
+
+static inline pending_ring_idx_t pending_index(unsigned i)
+{
+ return i & (MAX_PENDING_REQS-1);
+}
+
+static inline pending_ring_idx_t nr_pending_reqs(struct xen_netbk *netbk)
+{
+ return MAX_PENDING_REQS -
+ netbk->pending_prod + netbk->pending_cons;
+}
+
+static void xen_netbk_kick_thread(struct xen_netbk *netbk)
+{
+ wake_up(&netbk->wq);
+}
+
+/*
+ * Returns true if we should start a new receive buffer instead of
+ * adding 'size' bytes to a buffer which currently contains 'offset'
+ * bytes.
+ */
+static bool start_new_rx_buffer(int offset, unsigned long size, int head)
+{
+ /* simple case: we have completely filled the current buffer. */
+ if (offset == MAX_BUFFER_OFFSET)
+ return true;
+
+ /*
+ * complex case: start a fresh buffer if the current frag
+ * would overflow the current buffer but only if:
+ * (i) this frag would fit completely in the next buffer
+ * and (ii) there is already some data in the current buffer
+ * and (iii) this is not the head buffer.
+ *
+ * Where:
+ * - (i) stops us splitting a frag into two copies
+ * unless the frag is too large for a single buffer.
+ * - (ii) stops us from leaving a buffer pointlessly empty.
+ * - (iii) stops us leaving the first buffer
+ * empty. Strictly speaking this is already covered
+ * by (ii) but is explicitly checked because
+ * netfront relies on the first buffer being
+ * non-empty and can crash otherwise.
+ *
+ * This means we will effectively linearise small
+ * frags but do not needlessly split large buffers
+ * into multiple copies tend to give large frags their
+ * own buffers as before.
+ */
+ if ((offset + size > MAX_BUFFER_OFFSET) &&
+ (size <= MAX_BUFFER_OFFSET) && offset && !head)
+ return true;
+
+ return false;
+}
+
+/*
+ * Figure out how many ring slots we're going to need to send @skb to
+ * the guest. This function is essentially a dry run of
+ * netbk_gop_frag_copy.
+ */
+unsigned int xen_netbk_count_skb_slots(struct xenvif *vif, struct sk_buff *skb)
+{
+ unsigned int count;
+ int i, copy_off;
+
+ count = DIV_ROUND_UP(
+ offset_in_page(skb->data)+skb_headlen(skb), PAGE_SIZE);
+
+ copy_off = skb_headlen(skb) % PAGE_SIZE;
+
+ if (skb_shinfo(skb)->gso_size)
+ count++;
+
+ for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+ unsigned long size = skb_shinfo(skb)->frags[i].size;
+ unsigned long bytes;
+ while (size > 0) {
+ BUG_ON(copy_off > MAX_BUFFER_OFFSET);
+
+ if (start_new_rx_buffer(copy_off, size, 0)) {
+ count++;
+ copy_off = 0;
+ }
+
+ bytes = size;
+ if (copy_off + bytes > MAX_BUFFER_OFFSET)
+ bytes = MAX_BUFFER_OFFSET - copy_off;
+
+ copy_off += bytes;
+ size -= bytes;
+ }
+ }
+ return count;
+}
+
+struct netrx_pending_operations {
+ unsigned copy_prod, copy_cons;
+ unsigned meta_prod, meta_cons;
+ struct gnttab_copy *copy;
+ struct netbk_rx_meta *meta;
+ int copy_off;
+ grant_ref_t copy_gref;
+};
+
+static struct netbk_rx_meta *get_next_rx_buffer(struct xenvif *vif,
+ struct netrx_pending_operations *npo)
+{
+ struct netbk_rx_meta *meta;
+ struct xen_netif_rx_request *req;
+
+ req = RING_GET_REQUEST(&vif->rx, vif->rx.req_cons++);
+
+ meta = npo->meta + npo->meta_prod++;
+ meta->gso_size = 0;
+ meta->size = 0;
+ meta->id = req->id;
+
+ npo->copy_off = 0;
+ npo->copy_gref = req->gref;
+
+ return meta;
+}
+
+/*
+ * Set up the grant operations for this fragment. If it's a flipping
+ * interface, we also set up the unmap request from here.
+ */
+static void netbk_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
+ struct netrx_pending_operations *npo,
+ struct page *page, unsigned long size,
+ unsigned long offset, int *head)
+{
+ struct gnttab_copy *copy_gop;
+ struct netbk_rx_meta *meta;
+ /*
+ * These variables a used iff get_page_ext returns true,
+ * in which case they are guaranteed to be initialized.
+ */
+ unsigned int uninitialized_var(group), uninitialized_var(idx);
+ int foreign = get_page_ext(page, &group, &idx);
+ unsigned long bytes;
+
+ /* Data must not cross a page boundary. */
+ BUG_ON(size + offset > PAGE_SIZE);
+
+ meta = npo->meta + npo->meta_prod - 1;
+
+ while (size > 0) {
+ BUG_ON(npo->copy_off > MAX_BUFFER_OFFSET);
+
+ if (start_new_rx_buffer(npo->copy_off, size, *head)) {
+ /*
+ * Netfront requires there to be some data in the head
+ * buffer.
+ */
+ BUG_ON(*head);
+
+ meta = get_next_rx_buffer(vif, npo);
+ }
+
+ bytes = size;
+ if (npo->copy_off + bytes > MAX_BUFFER_OFFSET)
+ bytes = MAX_BUFFER_OFFSET - npo->copy_off;
+
+ copy_gop = npo->copy + npo->copy_prod++;
+ copy_gop->flags = GNTCOPY_dest_gref;
+ if (foreign) {
+ struct xen_netbk *netbk = &xen_netbk[group];
+ struct pending_tx_info *src_pend;
+
+ src_pend = &netbk->pending_tx_info[idx];
+
+ copy_gop->source.domid = src_pend->vif->domid;
+ copy_gop->source.u.ref = src_pend->req.gref;
+ copy_gop->flags |= GNTCOPY_source_gref;
+ } else {
+ void *vaddr = page_address(page);
+ copy_gop->source.domid = DOMID_SELF;
+ copy_gop->source.u.gmfn = virt_to_mfn(vaddr);
+ }
+ copy_gop->source.offset = offset;
+ copy_gop->dest.domid = vif->domid;
+
+ copy_gop->dest.offset = npo->copy_off;
+ copy_gop->dest.u.ref = npo->copy_gref;
+ copy_gop->len = bytes;
+
+ npo->copy_off += bytes;
+ meta->size += bytes;
+
+ offset += bytes;
+ size -= bytes;
+
+ /* Leave a gap for the GSO descriptor. */
+ if (*head && skb_shinfo(skb)->gso_size && !vif->gso_prefix)
+ vif->rx.req_cons++;
+
+ *head = 0; /* There must be something in this buffer now. */
+
+ }
+}
+
+/*
+ * Prepare an SKB to be transmitted to the frontend.
+ *
+ * This function is responsible for allocating grant operations, meta
+ * structures, etc.
+ *
+ * It returns the number of meta structures consumed. The number of
+ * ring slots used is always equal to the number of meta slots used
+ * plus the number of GSO descriptors used. Currently, we use either
+ * zero GSO descriptors (for non-GSO packets) or one descriptor (for
+ * frontend-side LRO).
+ */
+static int netbk_gop_skb(struct sk_buff *skb,
+ struct netrx_pending_operations *npo)
+{
+ struct xenvif *vif = netdev_priv(skb->dev);
+ int nr_frags = skb_shinfo(skb)->nr_frags;
+ int i;
+ struct xen_netif_rx_request *req;
+ struct netbk_rx_meta *meta;
+ unsigned char *data;
+ int head = 1;
+ int old_meta_prod;
+
+ old_meta_prod = npo->meta_prod;
+
+ /* Set up a GSO prefix descriptor, if necessary */
+ if (skb_shinfo(skb)->gso_size && vif->gso_prefix) {
+ req = RING_GET_REQUEST(&vif->rx, vif->rx.req_cons++);
+ meta = npo->meta + npo->meta_prod++;
+ meta->gso_size = skb_shinfo(skb)->gso_size;
+ meta->size = 0;
+ meta->id = req->id;
+ }
+
+ req = RING_GET_REQUEST(&vif->rx, vif->rx.req_cons++);
+ meta = npo->meta + npo->meta_prod++;
+
+ if (!vif->gso_prefix)
+ meta->gso_size = skb_shinfo(skb)->gso_size;
+ else
+ meta->gso_size = 0;
+
+ meta->size = 0;
+ meta->id = req->id;
+ npo->copy_off = 0;
+ npo->copy_gref = req->gref;
+
+ data = skb->data;
+ while (data < skb_tail_pointer(skb)) {
+ unsigned int offset = offset_in_page(data);
+ unsigned int len = PAGE_SIZE - offset;
+
+ if (data + len > skb_tail_pointer(skb))
+ len = skb_tail_pointer(skb) - data;
+
+ netbk_gop_frag_copy(vif, skb, npo,
+ virt_to_page(data), len, offset, &head);
+ data += len;
+ }
+
+ for (i = 0; i < nr_frags; i++) {
+ netbk_gop_frag_copy(vif, skb, npo,
+ skb_shinfo(skb)->frags[i].page,
+ skb_shinfo(skb)->frags[i].size,
+ skb_shinfo(skb)->frags[i].page_offset,
+ &head);
+ }
+
+ return npo->meta_prod - old_meta_prod;
+}
+
+/*
+ * This is a twin to netbk_gop_skb. Assume that netbk_gop_skb was
+ * used to set up the operations on the top of
+ * netrx_pending_operations, which have since been done. Check that
+ * they didn't give any errors and advance over them.
+ */
+static int netbk_check_gop(int nr_meta_slots, domid_t domid,
+ struct netrx_pending_operations *npo)
+{
+ struct gnttab_copy *copy_op;
+ int status = XEN_NETIF_RSP_OKAY;
+ int i;
+
+ for (i = 0; i < nr_meta_slots; i++) {
+ copy_op = npo->copy + npo->copy_cons++;
+ if (copy_op->status != GNTST_okay) {
+ pr_debug("Bad status %d from copy to DOM%d.\n",
+ copy_op->status, domid);
+ status = XEN_NETIF_RSP_ERROR;
+ }
+ }
+
+ return status;
+}
+
+static void netbk_add_frag_responses(struct xenvif *vif, int status,
+ struct netbk_rx_meta *meta,
+ int nr_meta_slots)
+{
+ int i;
+ unsigned long offset;
+
+ /* No fragments used */
+ if (nr_meta_slots <= 1)
+ return;
+
+ nr_meta_slots--;
+
+ for (i = 0; i < nr_meta_slots; i++) {
+ int flags;
+ if (i == nr_meta_slots - 1)
+ flags = 0;
+ else
+ flags = XEN_NETRXF_more_data;
+
+ offset = 0;
+ make_rx_response(vif, meta[i].id, status, offset,
+ meta[i].size, flags);
+ }
+}
+
+struct skb_cb_overlay {
+ int meta_slots_used;
+};
+
+static void xen_netbk_rx_action(struct xen_netbk *netbk)
+{
+ struct xenvif *vif = NULL;
+ s8 status;
+ u16 irq, flags;
+ struct xen_netif_rx_response *resp;
+ struct sk_buff_head rxq;
+ struct sk_buff *skb;
+ int notify_nr = 0;
+ int ret;
+ int nr_frags;
+ int count;
+ unsigned long offset;
+ struct skb_cb_overlay *sco;
+
+ struct netrx_pending_operations npo = {
+ .copy = netbk->grant_copy_op,
+ .meta = netbk->meta,
+ };
+
+ skb_queue_head_init(&rxq);
+
+ count = 0;
+
+ while ((skb = skb_dequeue(&netbk->rx_queue)) != NULL) {
+ vif = netdev_priv(skb->dev);
+ nr_frags = skb_shinfo(skb)->nr_frags;
+
+ sco = (struct skb_cb_overlay *)skb->cb;
+ sco->meta_slots_used = netbk_gop_skb(skb, &npo);
+
+ count += nr_frags + 1;
+
+ __skb_queue_tail(&rxq, skb);
+
+ /* Filled the batch queue? */
+ if (count + MAX_SKB_FRAGS >= XEN_NETIF_RX_RING_SIZE)
+ break;
+ }
+
+ BUG_ON(npo.meta_prod > ARRAY_SIZE(netbk->meta));
+
+ if (!npo.copy_prod)
+ return;
+
+ BUG_ON(npo.copy_prod > ARRAY_SIZE(netbk->grant_copy_op));
+ ret = HYPERVISOR_grant_table_op(GNTTABOP_copy, &netbk->grant_copy_op,
+ npo.copy_prod);
+ BUG_ON(ret != 0);
+
+ while ((skb = __skb_dequeue(&rxq)) != NULL) {
+ sco = (struct skb_cb_overlay *)skb->cb;
+
+ vif = netdev_priv(skb->dev);
+
+ if (netbk->meta[npo.meta_cons].gso_size && vif->gso_prefix) {
+ resp = RING_GET_RESPONSE(&vif->rx,
+ vif->rx.rsp_prod_pvt++);
+
+ resp->flags = XEN_NETRXF_gso_prefix | XEN_NETRXF_more_data;
+
+ resp->offset = netbk->meta[npo.meta_cons].gso_size;
+ resp->id = netbk->meta[npo.meta_cons].id;
+ resp->status = sco->meta_slots_used;
+
+ npo.meta_cons++;
+ sco->meta_slots_used--;
+ }
+
+
+ vif->stats.tx_bytes += skb->len;
+ vif->stats.tx_packets++;
+
+ status = netbk_check_gop(sco->meta_slots_used,
+ vif->domid, &npo);
+
+ if (sco->meta_slots_used == 1)
+ flags = 0;
+ else
+ flags = XEN_NETRXF_more_data;
+
+ if (skb->ip_summed == CHECKSUM_PARTIAL) /* local packet? */
+ flags |= XEN_NETRXF_csum_blank | XEN_NETRXF_data_validated;
+ else if (skb->ip_summed == CHECKSUM_UNNECESSARY)
+ /* remote but checksummed. */
+ flags |= XEN_NETRXF_data_validated;
+
+ offset = 0;
+ resp = make_rx_response(vif, netbk->meta[npo.meta_cons].id,
+ status, offset,
+ netbk->meta[npo.meta_cons].size,
+ flags);
+
+ if (netbk->meta[npo.meta_cons].gso_size && !vif->gso_prefix) {
+ struct xen_netif_extra_info *gso =
+ (struct xen_netif_extra_info *)
+ RING_GET_RESPONSE(&vif->rx,
+ vif->rx.rsp_prod_pvt++);
+
+ resp->flags |= XEN_NETRXF_extra_info;
+
+ gso->u.gso.size = netbk->meta[npo.meta_cons].gso_size;
+ gso->u.gso.type = XEN_NETIF_GSO_TYPE_TCPV4;
+ gso->u.gso.pad = 0;
+ gso->u.gso.features = 0;
+
+ gso->type = XEN_NETIF_EXTRA_TYPE_GSO;
+ gso->flags = 0;
+ }
+
+ netbk_add_frag_responses(vif, status,
+ netbk->meta + npo.meta_cons + 1,
+ sco->meta_slots_used);
+
+ RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&vif->rx, ret);
+ irq = vif->irq;
+ if (ret && !netbk->rx_notify[irq]) {
+ netbk->rx_notify[irq] = 1;
+ netbk->notify_list[notify_nr++] = irq;
+ }
+
+ if (netif_queue_stopped(vif->dev) &&
+ xenvif_schedulable(vif) &&
+ !xenvif_queue_full(vif))
+ netif_wake_queue(vif->dev);
+
+ xenvif_put(vif);
+ npo.meta_cons += sco->meta_slots_used;
+ dev_kfree_skb(skb);
+ }
+
+ while (notify_nr != 0) {
+ irq = netbk->notify_list[--notify_nr];
+ netbk->rx_notify[irq] = 0;
+ notify_remote_via_irq(irq);
+ }
+
+ /* More work to do? */
+ if (!skb_queue_empty(&netbk->rx_queue) &&
+ !timer_pending(&netbk->net_timer))
+ xen_netbk_kick_thread(netbk);
+}
+
+void xen_netbk_queue_tx_skb(struct xenvif *vif, struct sk_buff *skb)
+{
+ struct xen_netbk *netbk = vif->netbk;
+
+ skb_queue_tail(&netbk->rx_queue, skb);
+
+ xen_netbk_kick_thread(netbk);
+}
+
+static void xen_netbk_alarm(unsigned long data)
+{
+ struct xen_netbk *netbk = (struct xen_netbk *)data;
+ xen_netbk_kick_thread(netbk);
+}
+
+static int __on_net_schedule_list(struct xenvif *vif)
+{
+ return !list_empty(&vif->list);
+}
+
+/* Must be called with net_schedule_list_lock held */
+static void remove_from_net_schedule_list(struct xenvif *vif)
+{
+ if (likely(__on_net_schedule_list(vif))) {
+ list_del_init(&vif->list);
+ xenvif_put(vif);
+ }
+}
+
+static struct xenvif *poll_net_schedule_list(struct xen_netbk *netbk)
+{
+ struct xenvif *vif = NULL;
+
+ spin_lock_irq(&netbk->net_schedule_list_lock);
+ if (list_empty(&netbk->net_schedule_list))
+ goto out;
+
+ vif = list_first_entry(&netbk->net_schedule_list,
+ struct xenvif, list);
+ if (!vif)
+ goto out;
+
+ xenvif_get(vif);
+
+ remove_from_net_schedule_list(vif);
+out:
+ spin_unlock_irq(&netbk->net_schedule_list_lock);
+ return vif;
+}
+
+void xen_netbk_schedule_xenvif(struct xenvif *vif)
+{
+ unsigned long flags;
+
+ struct xen_netbk *netbk = vif->netbk;
+ if (__on_net_schedule_list(vif))
+ goto kick;
+
+ spin_lock_irqsave(&netbk->net_schedule_list_lock, flags);
+ if (!__on_net_schedule_list(vif) &&
+ likely(xenvif_schedulable(vif))) {
+ list_add_tail(&vif->list, &netbk->net_schedule_list);
+ xenvif_get(vif);
+ }
+ spin_unlock_irqrestore(&netbk->net_schedule_list_lock, flags);
+
+kick:
+ smp_mb();
+ if ((nr_pending_reqs(netbk) < (MAX_PENDING_REQS/2)) &&
+ !list_empty(&netbk->net_schedule_list))
+ xen_netbk_kick_thread(netbk);
+}
+
+void xen_netbk_deschedule_xenfif(struct xenvif *vif)
+{
+ struct xen_netbk *netbk = vif->netbk;
+ spin_lock_irq(&netbk->net_schedule_list_lock);
+ remove_from_net_schedule_list(vif);
+ spin_unlock_irq(&netbk->net_schedule_list_lock);
+}
+
+static void tx_add_credit(struct xenvif *vif)
+{
+ unsigned long max_burst, max_credit;
+
+ /*
+ * Allow a burst big enough to transmit a jumbo packet of up to 128kB.
+ * Otherwise the interface can seize up due to insufficient credit.
+ */
+ max_burst = RING_GET_REQUEST(&vif->tx, vif->tx.req_cons)->size;
+ max_burst = min(max_burst, 131072UL);
+ max_burst = max(max_burst, vif->credit_bytes);
+
+ /* Take care that adding a new chunk of credit doesn't wrap to zero. */
+ max_credit = vif->remaining_credit + vif->credit_bytes;
+ if (max_credit < vif->remaining_credit)
+ max_credit = ULONG_MAX; /* wrapped: clamp to ULONG_MAX */
+
+ vif->remaining_credit = min(max_credit, max_burst);
+}
+
+static void tx_credit_callback(unsigned long data)
+{
+ struct xenvif *vif = (struct xenvif *)data;
+ tx_add_credit(vif);
+ xenvif_schedule_work(vif);
+}
+
+static void netbk_tx_err(struct xenvif *vif,
+ struct xen_netif_tx_request *txp, RING_IDX end)
+{
+ RING_IDX cons = vif->tx.req_cons;
+
+ do {
+ make_tx_response(vif, txp, XEN_NETIF_RSP_ERROR);
+ if (cons >= end)
+ break;
+ txp = RING_GET_REQUEST(&vif->tx, cons++);
+ } while (1);
+ vif->tx.req_cons = cons;
+ xenvif_schedule_work(vif);
+ xenvif_put(vif);
+}
+
+static int netbk_count_requests(struct xenvif *vif,
+ struct xen_netif_tx_request *first,
+ struct xen_netif_tx_request *txp,
+ int work_to_do)
+{
+ RING_IDX cons = vif->tx.req_cons;
+ int frags = 0;
+
+ if (!(first->flags & XEN_NETTXF_more_data))
+ return 0;
+
+ do {
+ if (frags >= work_to_do) {
+ pr_debug("Need more frags\n");
+ return -frags;
+ }
+
+ if (unlikely(frags >= MAX_SKB_FRAGS)) {
+ pr_debug("Too many frags\n");
+ return -frags;
+ }
+
+ memcpy(txp, RING_GET_REQUEST(&vif->tx, cons + frags),
+ sizeof(*txp));
+ if (txp->size > first->size) {
+ pr_debug("Frags galore\n");
+ return -frags;
+ }
+
+ first->size -= txp->size;
+ frags++;
+
+ if (unlikely((txp->offset + txp->size) > PAGE_SIZE)) {
+ pr_debug("txp->offset: %x, size: %u\n",
+ txp->offset, txp->size);
+ return -frags;
+ }
+ } while ((txp++)->flags & XEN_NETTXF_more_data);
+ return frags;
+}
+
+static struct page *xen_netbk_alloc_page(struct xen_netbk *netbk,
+ struct sk_buff *skb,
+ unsigned long pending_idx)
+{
+ struct page *page;
+ page = alloc_page(GFP_KERNEL|__GFP_COLD);
+ if (!page)
+ return NULL;
+ set_page_ext(page, netbk, pending_idx);
+ netbk->mmap_pages[pending_idx] = page;
+ return page;
+}
+
+static struct gnttab_copy *xen_netbk_get_requests(struct xen_netbk *netbk,
+ struct xenvif *vif,
+ struct sk_buff *skb,
+ struct xen_netif_tx_request *txp,
+ struct gnttab_copy *gop)
+{
+ struct skb_shared_info *shinfo = skb_shinfo(skb);
+ skb_frag_t *frags = shinfo->frags;
+ unsigned long pending_idx = *((u16 *)skb->data);
+ int i, start;
+
+ /* Skip first skb fragment if it is on same page as header fragment. */
+ start = ((unsigned long)shinfo->frags[0].page == pending_idx);
+
+ for (i = start; i < shinfo->nr_frags; i++, txp++) {
+ struct page *page;
+ pending_ring_idx_t index;
+ struct pending_tx_info *pending_tx_info =
+ netbk->pending_tx_info;
+
+ index = pending_index(netbk->pending_cons++);
+ pending_idx = netbk->pending_ring[index];
+ page = xen_netbk_alloc_page(netbk, skb, pending_idx);
+ if (!page)
+ return NULL;
+
+ netbk->mmap_pages[pending_idx] = page;
+
+ gop->source.u.ref = txp->gref;
+ gop->source.domid = vif->domid;
+ gop->source.offset = txp->offset;
+
+ gop->dest.u.gmfn = virt_to_mfn(page_address(page));
+ gop->dest.domid = DOMID_SELF;
+ gop->dest.offset = txp->offset;
+
+ gop->len = txp->size;
+ gop->flags = GNTCOPY_source_gref;
+
+ gop++;
+
+ memcpy(&pending_tx_info[pending_idx].req, txp, sizeof(*txp));
+ xenvif_get(vif);
+ pending_tx_info[pending_idx].vif = vif;
+ frags[i].page = (void *)pending_idx;
+ }
+
+ return gop;
+}
+
+static int xen_netbk_tx_check_gop(struct xen_netbk *netbk,
+ struct sk_buff *skb,
+ struct gnttab_copy **gopp)
+{
+ struct gnttab_copy *gop = *gopp;
+ int pending_idx = *((u16 *)skb->data);
+ struct pending_tx_info *pending_tx_info = netbk->pending_tx_info;
+ struct xenvif *vif = pending_tx_info[pending_idx].vif;
+ struct xen_netif_tx_request *txp;
+ struct skb_shared_info *shinfo = skb_shinfo(skb);
+ int nr_frags = shinfo->nr_frags;
+ int i, err, start;
+
+ /* Check status of header. */
+ err = gop->status;
+ if (unlikely(err)) {
+ pending_ring_idx_t index;
+ index = pending_index(netbk->pending_prod++);
+ txp = &pending_tx_info[pending_idx].req;
+ make_tx_response(vif, txp, XEN_NETIF_RSP_ERROR);
+ netbk->pending_ring[index] = pending_idx;
+ xenvif_put(vif);
+ }
+
+ /* Skip first skb fragment if it is on same page as header fragment. */
+ start = ((unsigned long)shinfo->frags[0].page == pending_idx);
+
+ for (i = start; i < nr_frags; i++) {
+ int j, newerr;
+ pending_ring_idx_t index;
+
+ pending_idx = (unsigned long)shinfo->frags[i].page;
+
+ /* Check error status: if okay then remember grant handle. */
+ newerr = (++gop)->status;
+ if (likely(!newerr)) {
+ /* Had a previous error? Invalidate this fragment. */
+ if (unlikely(err))
+ xen_netbk_idx_release(netbk, pending_idx);
+ continue;
+ }
+
+ /* Error on this fragment: respond to client with an error. */
+ txp = &netbk->pending_tx_info[pending_idx].req;
+ make_tx_response(vif, txp, XEN_NETIF_RSP_ERROR);
+ index = pending_index(netbk->pending_prod++);
+ netbk->pending_ring[index] = pending_idx;
+ xenvif_put(vif);
+
+ /* Not the first error? Preceding frags already invalidated. */
+ if (err)
+ continue;
+
+ /* First error: invalidate header and preceding fragments. */
+ pending_idx = *((u16 *)skb->data);
+ xen_netbk_idx_release(netbk, pending_idx);
+ for (j = start; j < i; j++) {
+ pending_idx = (unsigned long)shinfo->frags[i].page;
+ xen_netbk_idx_release(netbk, pending_idx);
+ }
+
+ /* Remember the error: invalidate all subsequent fragments. */
+ err = newerr;
+ }
+
+ *gopp = gop + 1;
+ return err;
+}
+
+static void xen_netbk_fill_frags(struct xen_netbk *netbk, struct sk_buff *skb)
+{
+ struct skb_shared_info *shinfo = skb_shinfo(skb);
+ int nr_frags = shinfo->nr_frags;
+ int i;
+
+ for (i = 0; i < nr_frags; i++) {
+ skb_frag_t *frag = shinfo->frags + i;
+ struct xen_netif_tx_request *txp;
+ unsigned long pending_idx;
+
+ pending_idx = (unsigned long)frag->page;
+
+ txp = &netbk->pending_tx_info[pending_idx].req;
+ frag->page = virt_to_page(idx_to_kaddr(netbk, pending_idx));
+ frag->size = txp->size;
+ frag->page_offset = txp->offset;
+
+ skb->len += txp->size;
+ skb->data_len += txp->size;
+ skb->truesize += txp->size;
+
+ /* Take an extra reference to offset xen_netbk_idx_release */
+ get_page(netbk->mmap_pages[pending_idx]);
+ xen_netbk_idx_release(netbk, pending_idx);
+ }
+}
+
+static int xen_netbk_get_extras(struct xenvif *vif,
+ struct xen_netif_extra_info *extras,
+ int work_to_do)
+{
+ struct xen_netif_extra_info extra;
+ RING_IDX cons = vif->tx.req_cons;
+
+ do {
+ if (unlikely(work_to_do-- <= 0)) {
+ pr_debug("Missing extra info\n");
+ return -EBADR;
+ }
+
+ memcpy(&extra, RING_GET_REQUEST(&vif->tx, cons),
+ sizeof(extra));
+ if (unlikely(!extra.type ||
+ extra.type >= XEN_NETIF_EXTRA_TYPE_MAX)) {
+ vif->tx.req_cons = ++cons;
+ pr_debug("Invalid extra type: %d\n", extra.type);
+ return -EINVAL;
+ }
+
+ memcpy(&extras[extra.type - 1], &extra, sizeof(extra));
+ vif->tx.req_cons = ++cons;
+ } while (extra.flags & XEN_NETIF_EXTRA_FLAG_MORE);
+
+ return work_to_do;
+}
+
+static int netbk_set_skb_gso(struct sk_buff *skb,
+ struct xen_netif_extra_info *gso)
+{
+ if (!gso->u.gso.size) {
+ pr_debug("GSO size must not be zero.\n");
+ return -EINVAL;
+ }
+
+ /* Currently only TCPv4 S.O. is supported. */
+ if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4) {
+ pr_debug("Bad GSO type %d.\n", gso->u.gso.type);
+ return -EINVAL;
+ }
+
+ skb_shinfo(skb)->gso_size = gso->u.gso.size;
+ skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+
+ /* Header must be checked, and gso_segs computed. */
+ skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
+ skb_shinfo(skb)->gso_segs = 0;
+
+ return 0;
+}
+
+static int checksum_setup(struct xenvif *vif, struct sk_buff *skb)
+{
+ struct iphdr *iph;
+ unsigned char *th;
+ int err = -EPROTO;
+ int recalculate_partial_csum = 0;
+
+ /*
+ * A GSO SKB must be CHECKSUM_PARTIAL. However some buggy
+ * peers can fail to set NETRXF_csum_blank when sending a GSO
+ * frame. In this case force the SKB to CHECKSUM_PARTIAL and
+ * recalculate the partial checksum.
+ */
+ if (skb->ip_summed != CHECKSUM_PARTIAL && skb_is_gso(skb)) {
+ vif->rx_gso_checksum_fixup++;
+ skb->ip_summed = CHECKSUM_PARTIAL;
+ recalculate_partial_csum = 1;
+ }
+
+ /* A non-CHECKSUM_PARTIAL SKB does not require setup. */
+ if (skb->ip_summed != CHECKSUM_PARTIAL)
+ return 0;
+
+ if (skb->protocol != htons(ETH_P_IP))
+ goto out;
+
+ iph = (void *)skb->data;
+ th = skb->data + 4 * iph->ihl;
+ if (th >= skb_tail_pointer(skb))
+ goto out;
+
+ skb->csum_start = th - skb->head;
+ switch (iph->protocol) {
+ case IPPROTO_TCP:
+ skb->csum_offset = offsetof(struct tcphdr, check);
+
+ if (recalculate_partial_csum) {
+ struct tcphdr *tcph = (struct tcphdr *)th;
+ tcph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
+ skb->len - iph->ihl*4,
+ IPPROTO_TCP, 0);
+ }
+ break;
+ case IPPROTO_UDP:
+ skb->csum_offset = offsetof(struct udphdr, check);
+
+ if (recalculate_partial_csum) {
+ struct udphdr *udph = (struct udphdr *)th;
+ udph->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
+ skb->len - iph->ihl*4,
+ IPPROTO_UDP, 0);
+ }
+ break;
+ default:
+ if (net_ratelimit())
+ printk(KERN_ERR "Attempting to checksum a non-"
+ "TCP/UDP packet, dropping a protocol"
+ " %d packet", iph->protocol);
+ goto out;
+ }
+
+ if ((th + skb->csum_offset + 2) > skb_tail_pointer(skb))
+ goto out;
+
+ err = 0;
+
+out:
+ return err;
+}
+
+static bool tx_credit_exceeded(struct xenvif *vif, unsigned size)
+{
+ unsigned long now = jiffies;
+ unsigned long next_credit =
+ vif->credit_timeout.expires +
+ msecs_to_jiffies(vif->credit_usec / 1000);
+
+ /* Timer could already be pending in rare cases. */
+ if (timer_pending(&vif->credit_timeout))
+ return true;
+
+ /* Passed the point where we can replenish credit? */
+ if (time_after_eq(now, next_credit)) {
+ vif->credit_timeout.expires = now;
+ tx_add_credit(vif);
+ }
+
+ /* Still too big to send right now? Set a callback. */
+ if (size > vif->remaining_credit) {
+ vif->credit_timeout.data =
+ (unsigned long)vif;
+ vif->credit_timeout.function =
+ tx_credit_callback;
+ mod_timer(&vif->credit_timeout,
+ next_credit);
+
+ return true;
+ }
+
+ return false;
+}
+
+static unsigned xen_netbk_tx_build_gops(struct xen_netbk *netbk)
+{
+ struct gnttab_copy *gop = netbk->tx_copy_ops, *request_gop;
+ struct sk_buff *skb;
+ int ret;
+
+ while (((nr_pending_reqs(netbk) + MAX_SKB_FRAGS) < MAX_PENDING_REQS) &&
+ !list_empty(&netbk->net_schedule_list)) {
+ struct xenvif *vif;
+ struct xen_netif_tx_request txreq;
+ struct xen_netif_tx_request txfrags[MAX_SKB_FRAGS];
+ struct page *page;
+ struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX-1];
+ u16 pending_idx;
+ RING_IDX idx;
+ int work_to_do;
+ unsigned int data_len;
+ pending_ring_idx_t index;
+
+ /* Get a netif from the list with work to do. */
+ vif = poll_net_schedule_list(netbk);
+ if (!vif)
+ continue;
+
+ RING_FINAL_CHECK_FOR_REQUESTS(&vif->tx, work_to_do);
+ if (!work_to_do) {
+ xenvif_put(vif);
+ continue;
+ }
+
+ idx = vif->tx.req_cons;
+ rmb(); /* Ensure that we see the request before we copy it. */
+ memcpy(&txreq, RING_GET_REQUEST(&vif->tx, idx), sizeof(txreq));
+
+ /* Credit-based scheduling. */
+ if (txreq.size > vif->remaining_credit &&
+ tx_credit_exceeded(vif, txreq.size)) {
+ xenvif_put(vif);
+ continue;
+ }
+
+ vif->remaining_credit -= txreq.size;
+
+ work_to_do--;
+ vif->tx.req_cons = ++idx;
+
+ memset(extras, 0, sizeof(extras));
+ if (txreq.flags & XEN_NETTXF_extra_info) {
+ work_to_do = xen_netbk_get_extras(vif, extras,
+ work_to_do);
+ idx = vif->tx.req_cons;
+ if (unlikely(work_to_do < 0)) {
+ netbk_tx_err(vif, &txreq, idx);
+ continue;
+ }
+ }
+
+ ret = netbk_count_requests(vif, &txreq, txfrags, work_to_do);
+ if (unlikely(ret < 0)) {
+ netbk_tx_err(vif, &txreq, idx - ret);
+ continue;
+ }
+ idx += ret;
+
+ if (unlikely(txreq.size < ETH_HLEN)) {
+ pr_debug("Bad packet size: %d\n", txreq.size);
+ netbk_tx_err(vif, &txreq, idx);
+ continue;
+ }
+
+ /* No crossing a page as the payload mustn't fragment. */
+ if (unlikely((txreq.offset + txreq.size) > PAGE_SIZE)) {
+ pr_debug("txreq.offset: %x, size: %u, end: %lu\n",
+ txreq.offset, txreq.size,
+ (txreq.offset&~PAGE_MASK) + txreq.size);
+ netbk_tx_err(vif, &txreq, idx);
+ continue;
+ }
+
+ index = pending_index(netbk->pending_cons);
+ pending_idx = netbk->pending_ring[index];
+
+ data_len = (txreq.size > PKT_PROT_LEN &&
+ ret < MAX_SKB_FRAGS) ?
+ PKT_PROT_LEN : txreq.size;
+
+ skb = alloc_skb(data_len + NET_SKB_PAD + NET_IP_ALIGN,
+ GFP_ATOMIC | __GFP_NOWARN);
+ if (unlikely(skb == NULL)) {
+ pr_debug("Can't allocate a skb in start_xmit.\n");
+ netbk_tx_err(vif, &txreq, idx);
+ break;
+ }
+
+ /* Packets passed to netif_rx() must have some headroom. */
+ skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN);
+
+ if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) {
+ struct xen_netif_extra_info *gso;
+ gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1];
+
+ if (netbk_set_skb_gso(skb, gso)) {
+ kfree_skb(skb);
+ netbk_tx_err(vif, &txreq, idx);
+ continue;
+ }
+ }
+
+ /* XXX could copy straight to head */
+ page = xen_netbk_alloc_page(netbk, skb, pending_idx);
+ if (!page) {
+ kfree_skb(skb);
+ netbk_tx_err(vif, &txreq, idx);
+ continue;
+ }
+
+ netbk->mmap_pages[pending_idx] = page;
+
+ gop->source.u.ref = txreq.gref;
+ gop->source.domid = vif->domid;
+ gop->source.offset = txreq.offset;
+
+ gop->dest.u.gmfn = virt_to_mfn(page_address(page));
+ gop->dest.domid = DOMID_SELF;
+ gop->dest.offset = txreq.offset;
+
+ gop->len = txreq.size;
+ gop->flags = GNTCOPY_source_gref;
+
+ gop++;
+
+ memcpy(&netbk->pending_tx_info[pending_idx].req,
+ &txreq, sizeof(txreq));
+ netbk->pending_tx_info[pending_idx].vif = vif;
+ *((u16 *)skb->data) = pending_idx;
+
+ __skb_put(skb, data_len);
+
+ skb_shinfo(skb)->nr_frags = ret;
+ if (data_len < txreq.size) {
+ skb_shinfo(skb)->nr_frags++;
+ skb_shinfo(skb)->frags[0].page =
+ (void *)(unsigned long)pending_idx;
+ } else {
+ /* Discriminate from any valid pending_idx value. */
+ skb_shinfo(skb)->frags[0].page = (void *)~0UL;
+ }
+
+ __skb_queue_tail(&netbk->tx_queue, skb);
+
+ netbk->pending_cons++;
+
+ request_gop = xen_netbk_get_requests(netbk, vif,
+ skb, txfrags, gop);
+ if (request_gop == NULL) {
+ kfree_skb(skb);
+ netbk_tx_err(vif, &txreq, idx);
+ continue;
+ }
+ gop = request_gop;
+
+ vif->tx.req_cons = idx;
+ xenvif_schedule_work(vif);
+
+ if ((gop-netbk->tx_copy_ops) >= ARRAY_SIZE(netbk->tx_copy_ops))
+ break;
+ }
+
+ return gop - netbk->tx_copy_ops;
+}
+
+static void xen_netbk_tx_submit(struct xen_netbk *netbk)
+{
+ struct gnttab_copy *gop = netbk->tx_copy_ops;
+ struct sk_buff *skb;
+
+ while ((skb = __skb_dequeue(&netbk->tx_queue)) != NULL) {
+ struct xen_netif_tx_request *txp;
+ struct xenvif *vif;
+ u16 pending_idx;
+ unsigned data_len;
+
+ pending_idx = *((u16 *)skb->data);
+ vif = netbk->pending_tx_info[pending_idx].vif;
+ txp = &netbk->pending_tx_info[pending_idx].req;
+
+ /* Check the remap error code. */
+ if (unlikely(xen_netbk_tx_check_gop(netbk, skb, &gop))) {
+ pr_debug("netback grant failed.\n");
+ skb_shinfo(skb)->nr_frags = 0;
+ kfree_skb(skb);
+ continue;
+ }
+
+ data_len = skb->len;
+ memcpy(skb->data,
+ (void *)(idx_to_kaddr(netbk, pending_idx)|txp->offset),
+ data_len);
+ if (data_len < txp->size) {
+ /* Append the packet payload as a fragment. */
+ txp->offset += data_len;
+ txp->size -= data_len;
+ } else {
+ /* Schedule a response immediately. */
+ xen_netbk_idx_release(netbk, pending_idx);
+ }
+
+ if (txp->flags & XEN_NETTXF_csum_blank)
+ skb->ip_summed = CHECKSUM_PARTIAL;
+ else if (txp->flags & XEN_NETTXF_data_validated)
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+ xen_netbk_fill_frags(netbk, skb);
+
+ /*
+ * If the initial fragment was < PKT_PROT_LEN then
+ * pull through some bytes from the other fragments to
+ * increase the linear region to PKT_PROT_LEN bytes.
+ */
+ if (skb_headlen(skb) < PKT_PROT_LEN && skb_is_nonlinear(skb)) {
+ int target = min_t(int, skb->len, PKT_PROT_LEN);
+ __pskb_pull_tail(skb, target - skb_headlen(skb));
+ }
+
+ skb->dev = vif->dev;
+ skb->protocol = eth_type_trans(skb, skb->dev);
+
+ if (checksum_setup(vif, skb)) {
+ pr_debug("Can't setup checksum in net_tx_action\n");
+ kfree_skb(skb);
+ continue;
+ }
+
+ vif->stats.rx_bytes += skb->len;
+ vif->stats.rx_packets++;
+
+ netif_rx_ni(skb);
+ vif->dev->last_rx = jiffies;
+ }
+}
+
+/* Called after netfront has transmitted */
+static void xen_netbk_tx_action(struct xen_netbk *netbk)
+{
+ unsigned nr_gops;
+ int ret;
+
+ nr_gops = xen_netbk_tx_build_gops(netbk);
+
+ if (nr_gops == 0)
+ return;
+ ret = HYPERVISOR_grant_table_op(GNTTABOP_copy,
+ netbk->tx_copy_ops, nr_gops);
+ BUG_ON(ret);
+
+ xen_netbk_tx_submit(netbk);
+
+}
+
+static void xen_netbk_idx_release(struct xen_netbk *netbk, u16 pending_idx)
+{
+ struct xenvif *vif;
+ struct pending_tx_info *pending_tx_info;
+ pending_ring_idx_t index;
+
+ /* Already complete? */
+ if (netbk->mmap_pages[pending_idx] == NULL)
+ return;
+
+ pending_tx_info = &netbk->pending_tx_info[pending_idx];
+
+ vif = pending_tx_info->vif;
+
+ make_tx_response(vif, &pending_tx_info->req, XEN_NETIF_RSP_OKAY);
+
+ index = pending_index(netbk->pending_prod++);
+ netbk->pending_ring[index] = pending_idx;
+
+ xenvif_put(vif);
+
+ netbk->mmap_pages[pending_idx]->mapping = 0;
+ put_page(netbk->mmap_pages[pending_idx]);
+ netbk->mmap_pages[pending_idx] = NULL;
+}
+
+static void make_tx_response(struct xenvif *vif,
+ struct xen_netif_tx_request *txp,
+ s8 st)
+{
+ RING_IDX i = vif->tx.rsp_prod_pvt;
+ struct xen_netif_tx_response *resp;
+ int notify;
+
+ resp = RING_GET_RESPONSE(&vif->tx, i);
+ resp->id = txp->id;
+ resp->status = st;
+
+ if (txp->flags & XEN_NETTXF_extra_info)
+ RING_GET_RESPONSE(&vif->tx, ++i)->status = XEN_NETIF_RSP_NULL;
+
+ vif->tx.rsp_prod_pvt = ++i;
+ RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&vif->tx, notify);
+ if (notify)
+ notify_remote_via_irq(vif->irq);
+}
+
+static struct xen_netif_rx_response *make_rx_response(struct xenvif *vif,
+ u16 id,
+ s8 st,
+ u16 offset,
+ u16 size,
+ u16 flags)
+{
+ RING_IDX i = vif->rx.rsp_prod_pvt;
+ struct xen_netif_rx_response *resp;
+
+ resp = RING_GET_RESPONSE(&vif->rx, i);
+ resp->offset = offset;
+ resp->flags = flags;
+ resp->id = id;
+ resp->status = (s16)size;
+ if (st < 0)
+ resp->status = (s16)st;
+
+ vif->rx.rsp_prod_pvt = ++i;
+
+ return resp;
+}
+
+static inline int rx_work_todo(struct xen_netbk *netbk)
+{
+ return !skb_queue_empty(&netbk->rx_queue);
+}
+
+static inline int tx_work_todo(struct xen_netbk *netbk)
+{
+
+ if (((nr_pending_reqs(netbk) + MAX_SKB_FRAGS) < MAX_PENDING_REQS) &&
+ !list_empty(&netbk->net_schedule_list))
+ return 1;
+
+ return 0;
+}
+
+static int xen_netbk_kthread(void *data)
+{
+ struct xen_netbk *netbk = (struct xen_netbk *)data;
+ while (!kthread_should_stop()) {
+ wait_event_interruptible(netbk->wq,
+ rx_work_todo(netbk)
+ || tx_work_todo(netbk)
+ || kthread_should_stop());
+ cond_resched();
+
+ if (kthread_should_stop())
+ break;
+
+ if (rx_work_todo(netbk))
+ xen_netbk_rx_action(netbk);
+
+ if (tx_work_todo(netbk))
+ xen_netbk_tx_action(netbk);
+ }
+
+ return 0;
+}
+
+static int __init netback_init(void)
+{
+ int i;
+ int rc = 0;
+ int group;
+
+ if (!xen_pv_domain())
+ return -ENODEV;
+
+ xen_netbk_group_nr = num_online_cpus();
+ xen_netbk = vmalloc(sizeof(struct xen_netbk) * xen_netbk_group_nr);
+ if (!xen_netbk) {
+ printk(KERN_ALERT "%s: out of memory\n", __func__);
+ return -ENOMEM;
+ }
+ memset(xen_netbk, 0, sizeof(struct xen_netbk) * xen_netbk_group_nr);
+
+ for (group = 0; group < xen_netbk_group_nr; group++) {
+ struct xen_netbk *netbk = &xen_netbk[group];
+ skb_queue_head_init(&netbk->rx_queue);
+ skb_queue_head_init(&netbk->tx_queue);
+
+ init_timer(&netbk->net_timer);
+ netbk->net_timer.data = (unsigned long)netbk;
+ netbk->net_timer.function = xen_netbk_alarm;
+
+ netbk->pending_cons = 0;
+ netbk->pending_prod = MAX_PENDING_REQS;
+ for (i = 0; i < MAX_PENDING_REQS; i++)
+ netbk->pending_ring[i] = i;
+
+ init_waitqueue_head(&netbk->wq);
+ netbk->task = kthread_create(xen_netbk_kthread,
+ (void *)netbk,
+ "netback/%u", group);
+
+ if (IS_ERR(netbk->task)) {
+ printk(KERN_ALERT "kthread_run() fails at netback\n");
+ del_timer(&netbk->net_timer);
+ rc = PTR_ERR(netbk->task);
+ goto failed_init;
+ }
+
+ kthread_bind(netbk->task, group);
+
+ INIT_LIST_HEAD(&netbk->net_schedule_list);
+
+ spin_lock_init(&netbk->net_schedule_list_lock);
+
+ atomic_set(&netbk->netfront_count, 0);
+
+ wake_up_process(netbk->task);
+ }
+
+ rc = xenvif_xenbus_init();
+ if (rc)
+ goto failed_init;
+
+ return 0;
+
+failed_init:
+ for (i = 0; i < group; i++) {
+ struct xen_netbk *netbk = &xen_netbk[i];
+ int j;
+ for (j = 0; j < MAX_PENDING_REQS; j++) {
+ if (netbk->mmap_pages[i])
+ __free_page(netbk->mmap_pages[i]);
+ }
+ del_timer(&netbk->net_timer);
+ kthread_stop(netbk->task);
+ }
+ vfree(xen_netbk);
+ return rc;
+
+}
+
+module_init(netback_init);
+
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/xen-netback/xenbus.c b/drivers/net/xen-netback/xenbus.c
new file mode 100644
index 0000000..22b8c35
--- /dev/null
+++ b/drivers/net/xen-netback/xenbus.c
@@ -0,0 +1,490 @@
+/*
+ * Xenbus code for netif backend
+ *
+ * Copyright (C) 2005 Rusty Russell <rusty@rustcorp.com.au>
+ * Copyright (C) 2005 XenSource Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+*/
+
+#include "common.h"
+
+struct backend_info {
+ struct xenbus_device *dev;
+ struct xenvif *vif;
+ enum xenbus_state frontend_state;
+ struct xenbus_watch hotplug_status_watch;
+ int have_hotplug_status_watch:1;
+};
+
+static int connect_rings(struct backend_info *);
+static void connect(struct backend_info *);
+static void backend_create_xenvif(struct backend_info *be);
+static void unregister_hotplug_status_watch(struct backend_info *be);
+
+static int netback_remove(struct xenbus_device *dev)
+{
+ struct backend_info *be = dev_get_drvdata(&dev->dev);
+
+ unregister_hotplug_status_watch(be);
+ if (be->vif) {
+ kobject_uevent(&dev->dev.kobj, KOBJ_OFFLINE);
+ xenbus_rm(XBT_NIL, dev->nodename, "hotplug-status");
+ xenvif_disconnect(be->vif);
+ be->vif = NULL;
+ }
+ kfree(be);
+ dev_set_drvdata(&dev->dev, NULL);
+ return 0;
+}
+
+
+/**
+ * Entry point to this code when a new device is created. Allocate the basic
+ * structures and switch to InitWait.
+ */
+static int netback_probe(struct xenbus_device *dev,
+ const struct xenbus_device_id *id)
+{
+ const char *message;
+ struct xenbus_transaction xbt;
+ int err;
+ int sg;
+ struct backend_info *be = kzalloc(sizeof(struct backend_info),
+ GFP_KERNEL);
+ if (!be) {
+ xenbus_dev_fatal(dev, -ENOMEM,
+ "allocating backend structure");
+ return -ENOMEM;
+ }
+
+ be->dev = dev;
+ dev_set_drvdata(&dev->dev, be);
+
+ sg = 1;
+
+ do {
+ err = xenbus_transaction_start(&xbt);
+ if (err) {
+ xenbus_dev_fatal(dev, err, "starting transaction");
+ goto fail;
+ }
+
+ err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", sg);
+ if (err) {
+ message = "writing feature-sg";
+ goto abort_transaction;
+ }
+
+ err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv4",
+ "%d", sg);
+ if (err) {
+ message = "writing feature-gso-tcpv4";
+ goto abort_transaction;
+ }
+
+ /* We support rx-copy path. */
+ err = xenbus_printf(xbt, dev->nodename,
+ "feature-rx-copy", "%d", 1);
+ if (err) {
+ message = "writing feature-rx-copy";
+ goto abort_transaction;
+ }
+
+ /*
+ * We don't support rx-flip path (except old guests who don't
+ * grok this feature flag).
+ */
+ err = xenbus_printf(xbt, dev->nodename,
+ "feature-rx-flip", "%d", 0);
+ if (err) {
+ message = "writing feature-rx-flip";
+ goto abort_transaction;
+ }
+
+ err = xenbus_transaction_end(xbt, 0);
+ } while (err == -EAGAIN);
+
+ if (err) {
+ xenbus_dev_fatal(dev, err, "completing transaction");
+ goto fail;
+ }
+
+ err = xenbus_switch_state(dev, XenbusStateInitWait);
+ if (err)
+ goto fail;
+
+ /* This kicks hotplug scripts, so do it immediately. */
+ backend_create_xenvif(be);
+
+ return 0;
+
+abort_transaction:
+ xenbus_transaction_end(xbt, 1);
+ xenbus_dev_fatal(dev, err, "%s", message);
+fail:
+ pr_debug("failed");
+ netback_remove(dev);
+ return err;
+}
+
+
+/*
+ * Handle the creation of the hotplug script environment. We add the script
+ * and vif variables to the environment, for the benefit of the vif-* hotplug
+ * scripts.
+ */
+static int netback_uevent(struct xenbus_device *xdev,
+ struct kobj_uevent_env *env)
+{
+ struct backend_info *be = dev_get_drvdata(&xdev->dev);
+ char *val;
+
+ val = xenbus_read(XBT_NIL, xdev->nodename, "script", NULL);
+ if (IS_ERR(val)) {
+ int err = PTR_ERR(val);
+ xenbus_dev_fatal(xdev, err, "reading script");
+ return err;
+ } else {
+ if (add_uevent_var(env, "script=%s", val)) {
+ kfree(val);
+ return -ENOMEM;
+ }
+ kfree(val);
+ }
+
+ if (!be || !be->vif)
+ return 0;
+
+ return add_uevent_var(env, "vif=%s", be->vif->dev->name);
+}
+
+
+static void backend_create_xenvif(struct backend_info *be)
+{
+ int err;
+ long handle;
+ struct xenbus_device *dev = be->dev;
+
+ if (be->vif != NULL)
+ return;
+
+ err = xenbus_scanf(XBT_NIL, dev->nodename, "handle", "%li", &handle);
+ if (err != 1) {
+ xenbus_dev_fatal(dev, err, "reading handle");
+ return;
+ }
+
+ be->vif = xenvif_alloc(&dev->dev, dev->otherend_id, handle);
+ if (IS_ERR(be->vif)) {
+ err = PTR_ERR(be->vif);
+ be->vif = NULL;
+ xenbus_dev_fatal(dev, err, "creating interface");
+ return;
+ }
+
+ kobject_uevent(&dev->dev.kobj, KOBJ_ONLINE);
+}
+
+
+static void disconnect_backend(struct xenbus_device *dev)
+{
+ struct backend_info *be = dev_get_drvdata(&dev->dev);
+
+ if (be->vif) {
+ xenbus_rm(XBT_NIL, dev->nodename, "hotplug-status");
+ xenvif_disconnect(be->vif);
+ be->vif = NULL;
+ }
+}
+
+/**
+ * Callback received when the frontend's state changes.
+ */
+static void frontend_changed(struct xenbus_device *dev,
+ enum xenbus_state frontend_state)
+{
+ struct backend_info *be = dev_get_drvdata(&dev->dev);
+
+ pr_debug("frontend state %s", xenbus_strstate(frontend_state));
+
+ be->frontend_state = frontend_state;
+
+ switch (frontend_state) {
+ case XenbusStateInitialising:
+ if (dev->state == XenbusStateClosed) {
+ printk(KERN_INFO "%s: %s: prepare for reconnect\n",
+ __func__, dev->nodename);
+ xenbus_switch_state(dev, XenbusStateInitWait);
+ }
+ break;
+
+ case XenbusStateInitialised:
+ break;
+
+ case XenbusStateConnected:
+ if (dev->state == XenbusStateConnected)
+ break;
+ backend_create_xenvif(be);
+ if (be->vif)
+ connect(be);
+ break;
+
+ case XenbusStateClosing:
+ if (be->vif)
+ kobject_uevent(&dev->dev.kobj, KOBJ_OFFLINE);
+ disconnect_backend(dev);
+ xenbus_switch_state(dev, XenbusStateClosing);
+ break;
+
+ case XenbusStateClosed:
+ xenbus_switch_state(dev, XenbusStateClosed);
+ if (xenbus_dev_is_online(dev))
+ break;
+ /* fall through if not online */
+ case XenbusStateUnknown:
+ device_unregister(&dev->dev);
+ break;
+
+ default:
+ xenbus_dev_fatal(dev, -EINVAL, "saw state %d at frontend",
+ frontend_state);
+ break;
+ }
+}
+
+
+static void xen_net_read_rate(struct xenbus_device *dev,
+ unsigned long *bytes, unsigned long *usec)
+{
+ char *s, *e;
+ unsigned long b, u;
+ char *ratestr;
+
+ /* Default to unlimited bandwidth. */
+ *bytes = ~0UL;
+ *usec = 0;
+
+ ratestr = xenbus_read(XBT_NIL, dev->nodename, "rate", NULL);
+ if (IS_ERR(ratestr))
+ return;
+
+ s = ratestr;
+ b = simple_strtoul(s, &e, 10);
+ if ((s == e) || (*e != ','))
+ goto fail;
+
+ s = e + 1;
+ u = simple_strtoul(s, &e, 10);
+ if ((s == e) || (*e != '\0'))
+ goto fail;
+
+ *bytes = b;
+ *usec = u;
+
+ kfree(ratestr);
+ return;
+
+ fail:
+ pr_warn("Failed to parse network rate limit. Traffic unlimited.\n");
+ kfree(ratestr);
+}
+
+static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[])
+{
+ char *s, *e, *macstr;
+ int i;
+
+ macstr = s = xenbus_read(XBT_NIL, dev->nodename, "mac", NULL);
+ if (IS_ERR(macstr))
+ return PTR_ERR(macstr);
+
+ for (i = 0; i < ETH_ALEN; i++) {
+ mac[i] = simple_strtoul(s, &e, 16);
+ if ((s == e) || (*e != ((i == ETH_ALEN-1) ? '\0' : ':'))) {
+ kfree(macstr);
+ return -ENOENT;
+ }
+ s = e+1;
+ }
+
+ kfree(macstr);
+ return 0;
+}
+
+static void unregister_hotplug_status_watch(struct backend_info *be)
+{
+ if (be->have_hotplug_status_watch) {
+ unregister_xenbus_watch(&be->hotplug_status_watch);
+ kfree(be->hotplug_status_watch.node);
+ }
+ be->have_hotplug_status_watch = 0;
+}
+
+static void hotplug_status_changed(struct xenbus_watch *watch,
+ const char **vec,
+ unsigned int vec_size)
+{
+ struct backend_info *be = container_of(watch,
+ struct backend_info,
+ hotplug_status_watch);
+ char *str;
+ unsigned int len;
+
+ str = xenbus_read(XBT_NIL, be->dev->nodename, "hotplug-status", &len);
+ if (IS_ERR(str))
+ return;
+ if (len == sizeof("connected")-1 && !memcmp(str, "connected", len)) {
+ xenbus_switch_state(be->dev, XenbusStateConnected);
+ /* Not interested in this watch anymore. */
+ unregister_hotplug_status_watch(be);
+ }
+ kfree(str);
+}
+
+static void connect(struct backend_info *be)
+{
+ int err;
+ struct xenbus_device *dev = be->dev;
+
+ err = connect_rings(be);
+ if (err)
+ return;
+
+ err = xen_net_read_mac(dev, be->vif->fe_dev_addr);
+ if (err) {
+ xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename);
+ return;
+ }
+
+ xen_net_read_rate(dev, &be->vif->credit_bytes,
+ &be->vif->credit_usec);
+ be->vif->remaining_credit = be->vif->credit_bytes;
+
+ unregister_hotplug_status_watch(be);
+ err = xenbus_watch_pathfmt(dev, &be->hotplug_status_watch,
+ hotplug_status_changed,
+ "%s/%s", dev->nodename, "hotplug-status");
+ if (err) {
+ /* Switch now, since we can't do a watch. */
+ xenbus_switch_state(dev, XenbusStateConnected);
+ } else {
+ be->have_hotplug_status_watch = 1;
+ }
+
+ netif_wake_queue(be->vif->dev);
+}
+
+
+static int connect_rings(struct backend_info *be)
+{
+ struct xenvif *vif = be->vif;
+ struct xenbus_device *dev = be->dev;
+ unsigned long tx_ring_ref, rx_ring_ref;
+ unsigned int evtchn, rx_copy;
+ int err;
+ int val;
+
+ err = xenbus_gather(XBT_NIL, dev->otherend,
+ "tx-ring-ref", "%lu", &tx_ring_ref,
+ "rx-ring-ref", "%lu", &rx_ring_ref,
+ "event-channel", "%u", &evtchn, NULL);
+ if (err) {
+ xenbus_dev_fatal(dev, err,
+ "reading %s/ring-ref and event-channel",
+ dev->otherend);
+ return err;
+ }
+
+ err = xenbus_scanf(XBT_NIL, dev->otherend, "request-rx-copy", "%u",
+ &rx_copy);
+ if (err == -ENOENT) {
+ err = 0;
+ rx_copy = 0;
+ }
+ if (err < 0) {
+ xenbus_dev_fatal(dev, err, "reading %s/request-rx-copy",
+ dev->otherend);
+ return err;
+ }
+ if (!rx_copy)
+ return -EOPNOTSUPP;
+
+ if (vif->dev->tx_queue_len != 0) {
+ if (xenbus_scanf(XBT_NIL, dev->otherend,
+ "feature-rx-notify", "%d", &val) < 0)
+ val = 0;
+ if (val)
+ vif->can_queue = 1;
+ else
+ /* Must be non-zero for pfifo_fast to work. */
+ vif->dev->tx_queue_len = 1;
+ }
+
+ if (xenbus_scanf(XBT_NIL, dev->otherend, "feature-sg",
+ "%d", &val) < 0)
+ val = 0;
+ vif->can_sg = !!val;
+
+ if (xenbus_scanf(XBT_NIL, dev->otherend, "feature-gso-tcpv4",
+ "%d", &val) < 0)
+ val = 0;
+ vif->gso = !!val;
+
+ if (xenbus_scanf(XBT_NIL, dev->otherend, "feature-gso-tcpv4-prefix",
+ "%d", &val) < 0)
+ val = 0;
+ vif->gso_prefix = !!val;
+
+ if (xenbus_scanf(XBT_NIL, dev->otherend, "feature-no-csum-offload",
+ "%d", &val) < 0)
+ val = 0;
+ vif->csum = !val;
+
+ /* Map the shared frame, irq etc. */
+ err = xenvif_connect(vif, tx_ring_ref, rx_ring_ref, evtchn);
+ if (err) {
+ xenbus_dev_fatal(dev, err,
+ "mapping shared-frames %lu/%lu port %u",
+ tx_ring_ref, rx_ring_ref, evtchn);
+ return err;
+ }
+ return 0;
+}
+
+
+/* ** Driver Registration ** */
+
+
+static const struct xenbus_device_id netback_ids[] = {
+ { "vif" },
+ { "" }
+};
+
+
+static struct xenbus_driver netback = {
+ .name = "vif",
+ .owner = THIS_MODULE,
+ .ids = netback_ids,
+ .probe = netback_probe,
+ .remove = netback_remove,
+ .uevent = netback_uevent,
+ .otherend_changed = frontend_changed,
+};
+
+int xenvif_xenbus_init(void)
+{
+ return xenbus_register_backend(&netback);
+}
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index 458bb57..cc23d42 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -356,7 +356,7 @@ static void xennet_tx_buf_gc(struct net_device *dev)
struct xen_netif_tx_response *txrsp;
txrsp = RING_GET_RESPONSE(&np->tx, cons);
- if (txrsp->status == NETIF_RSP_NULL)
+ if (txrsp->status == XEN_NETIF_RSP_NULL)
continue;
id = txrsp->id;
@@ -413,7 +413,7 @@ static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
larger than a page), split it it into page-sized chunks. */
while (len > PAGE_SIZE - offset) {
tx->size = PAGE_SIZE - offset;
- tx->flags |= NETTXF_more_data;
+ tx->flags |= XEN_NETTXF_more_data;
len -= tx->size;
data += tx->size;
offset = 0;
@@ -439,7 +439,7 @@ static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
for (i = 0; i < frags; i++) {
skb_frag_t *frag = skb_shinfo(skb)->frags + i;
- tx->flags |= NETTXF_more_data;
+ tx->flags |= XEN_NETTXF_more_data;
id = get_id_from_freelist(&np->tx_skb_freelist, np->tx_skbs);
np->tx_skbs[id].skb = skb_get(skb);
@@ -514,10 +514,10 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
tx->flags = 0;
if (skb->ip_summed == CHECKSUM_PARTIAL)
/* local packet? */
- tx->flags |= NETTXF_csum_blank | NETTXF_data_validated;
+ tx->flags |= XEN_NETTXF_csum_blank | XEN_NETTXF_data_validated;
else if (skb->ip_summed == CHECKSUM_UNNECESSARY)
/* remote but checksummed. */
- tx->flags |= NETTXF_data_validated;
+ tx->flags |= XEN_NETTXF_data_validated;
if (skb_shinfo(skb)->gso_size) {
struct xen_netif_extra_info *gso;
@@ -528,7 +528,7 @@ static int xennet_start_xmit(struct sk_buff *skb, struct net_device *dev)
if (extra)
extra->flags |= XEN_NETIF_EXTRA_FLAG_MORE;
else
- tx->flags |= NETTXF_extra_info;
+ tx->flags |= XEN_NETTXF_extra_info;
gso->u.gso.size = skb_shinfo(skb)->gso_size;
gso->u.gso.type = XEN_NETIF_GSO_TYPE_TCPV4;
@@ -648,7 +648,7 @@ static int xennet_get_responses(struct netfront_info *np,
int err = 0;
unsigned long ret;
- if (rx->flags & NETRXF_extra_info) {
+ if (rx->flags & XEN_NETRXF_extra_info) {
err = xennet_get_extras(np, extras, rp);
cons = np->rx.rsp_cons;
}
@@ -685,7 +685,7 @@ static int xennet_get_responses(struct netfront_info *np,
__skb_queue_tail(list, skb);
next:
- if (!(rx->flags & NETRXF_more_data))
+ if (!(rx->flags & XEN_NETRXF_more_data))
break;
if (cons + frags == rp) {
@@ -950,9 +950,9 @@ err:
skb->truesize += skb->data_len - (RX_COPY_THRESHOLD - len);
skb->len += skb->data_len;
- if (rx->flags & NETRXF_csum_blank)
+ if (rx->flags & XEN_NETRXF_csum_blank)
skb->ip_summed = CHECKSUM_PARTIAL;
- else if (rx->flags & NETRXF_data_validated)
+ else if (rx->flags & XEN_NETRXF_data_validated)
skb->ip_summed = CHECKSUM_UNNECESSARY;
__skb_queue_tail(&rxq, skb);
diff --git a/include/xen/interface/io/netif.h b/include/xen/interface/io/netif.h
index 518481c..cb94668 100644
--- a/include/xen/interface/io/netif.h
+++ b/include/xen/interface/io/netif.h
@@ -22,50 +22,50 @@
/*
* This is the 'wire' format for packets:
- * Request 1: netif_tx_request -- NETTXF_* (any flags)
- * [Request 2: netif_tx_extra] (only if request 1 has NETTXF_extra_info)
- * [Request 3: netif_tx_extra] (only if request 2 has XEN_NETIF_EXTRA_MORE)
- * Request 4: netif_tx_request -- NETTXF_more_data
- * Request 5: netif_tx_request -- NETTXF_more_data
+ * Request 1: xen_netif_tx_request -- XEN_NETTXF_* (any flags)
+ * [Request 2: xen_netif_extra_info] (only if request 1 has XEN_NETTXF_extra_info)
+ * [Request 3: xen_netif_extra_info] (only if request 2 has XEN_NETIF_EXTRA_MORE)
+ * Request 4: xen_netif_tx_request -- XEN_NETTXF_more_data
+ * Request 5: xen_netif_tx_request -- XEN_NETTXF_more_data
* ...
- * Request N: netif_tx_request -- 0
+ * Request N: xen_netif_tx_request -- 0
*/
/* Protocol checksum field is blank in the packet (hardware offload)? */
-#define _NETTXF_csum_blank (0)
-#define NETTXF_csum_blank (1U<<_NETTXF_csum_blank)
+#define _XEN_NETTXF_csum_blank (0)
+#define XEN_NETTXF_csum_blank (1U<<_XEN_NETTXF_csum_blank)
/* Packet data has been validated against protocol checksum. */
-#define _NETTXF_data_validated (1)
-#define NETTXF_data_validated (1U<<_NETTXF_data_validated)
+#define _XEN_NETTXF_data_validated (1)
+#define XEN_NETTXF_data_validated (1U<<_XEN_NETTXF_data_validated)
/* Packet continues in the next request descriptor. */
-#define _NETTXF_more_data (2)
-#define NETTXF_more_data (1U<<_NETTXF_more_data)
+#define _XEN_NETTXF_more_data (2)
+#define XEN_NETTXF_more_data (1U<<_XEN_NETTXF_more_data)
/* Packet to be followed by extra descriptor(s). */
-#define _NETTXF_extra_info (3)
-#define NETTXF_extra_info (1U<<_NETTXF_extra_info)
+#define _XEN_NETTXF_extra_info (3)
+#define XEN_NETTXF_extra_info (1U<<_XEN_NETTXF_extra_info)
struct xen_netif_tx_request {
grant_ref_t gref; /* Reference to buffer page */
uint16_t offset; /* Offset within buffer page */
- uint16_t flags; /* NETTXF_* */
+ uint16_t flags; /* XEN_NETTXF_* */
uint16_t id; /* Echoed in response message. */
uint16_t size; /* Packet size in bytes. */
};
-/* Types of netif_extra_info descriptors. */
-#define XEN_NETIF_EXTRA_TYPE_NONE (0) /* Never used - invalid */
-#define XEN_NETIF_EXTRA_TYPE_GSO (1) /* u.gso */
-#define XEN_NETIF_EXTRA_TYPE_MAX (2)
+/* Types of xen_netif_extra_info descriptors. */
+#define XEN_NETIF_EXTRA_TYPE_NONE (0) /* Never used - invalid */
+#define XEN_NETIF_EXTRA_TYPE_GSO (1) /* u.gso */
+#define XEN_NETIF_EXTRA_TYPE_MAX (2)
-/* netif_extra_info flags. */
-#define _XEN_NETIF_EXTRA_FLAG_MORE (0)
-#define XEN_NETIF_EXTRA_FLAG_MORE (1U<<_XEN_NETIF_EXTRA_FLAG_MORE)
+/* xen_netif_extra_info flags. */
+#define _XEN_NETIF_EXTRA_FLAG_MORE (0)
+#define XEN_NETIF_EXTRA_FLAG_MORE (1U<<_XEN_NETIF_EXTRA_FLAG_MORE)
/* GSO types - only TCPv4 currently supported. */
-#define XEN_NETIF_GSO_TYPE_TCPV4 (1)
+#define XEN_NETIF_GSO_TYPE_TCPV4 (1)
/*
* This structure needs to fit within both netif_tx_request and
@@ -107,7 +107,7 @@ struct xen_netif_extra_info {
struct xen_netif_tx_response {
uint16_t id;
- int16_t status; /* NETIF_RSP_* */
+ int16_t status; /* XEN_NETIF_RSP_* */
};
struct xen_netif_rx_request {
@@ -116,25 +116,29 @@ struct xen_netif_rx_request {
};
/* Packet data has been validated against protocol checksum. */
-#define _NETRXF_data_validated (0)
-#define NETRXF_data_validated (1U<<_NETRXF_data_validated)
+#define _XEN_NETRXF_data_validated (0)
+#define XEN_NETRXF_data_validated (1U<<_XEN_NETRXF_data_validated)
/* Protocol checksum field is blank in the packet (hardware offload)? */
-#define _NETRXF_csum_blank (1)
-#define NETRXF_csum_blank (1U<<_NETRXF_csum_blank)
+#define _XEN_NETRXF_csum_blank (1)
+#define XEN_NETRXF_csum_blank (1U<<_XEN_NETRXF_csum_blank)
/* Packet continues in the next request descriptor. */
-#define _NETRXF_more_data (2)
-#define NETRXF_more_data (1U<<_NETRXF_more_data)
+#define _XEN_NETRXF_more_data (2)
+#define XEN_NETRXF_more_data (1U<<_XEN_NETRXF_more_data)
/* Packet to be followed by extra descriptor(s). */
-#define _NETRXF_extra_info (3)
-#define NETRXF_extra_info (1U<<_NETRXF_extra_info)
+#define _XEN_NETRXF_extra_info (3)
+#define XEN_NETRXF_extra_info (1U<<_XEN_NETRXF_extra_info)
+
+/* GSO Prefix descriptor. */
+#define _XEN_NETRXF_gso_prefix (4)
+#define XEN_NETRXF_gso_prefix (1U<<_XEN_NETRXF_gso_prefix)
struct xen_netif_rx_response {
uint16_t id;
uint16_t offset; /* Offset in page of start of received packet */
- uint16_t flags; /* NETRXF_* */
+ uint16_t flags; /* XEN_NETRXF_* */
int16_t status; /* -ve: BLKIF_RSP_* ; +ve: Rx'ed pkt size. */
};
@@ -149,10 +153,10 @@ DEFINE_RING_TYPES(xen_netif_rx,
struct xen_netif_rx_request,
struct xen_netif_rx_response);
-#define NETIF_RSP_DROPPED -2
-#define NETIF_RSP_ERROR -1
-#define NETIF_RSP_OKAY 0
-/* No response: used for auxiliary requests (e.g., netif_tx_extra). */
-#define NETIF_RSP_NULL 1
+#define XEN_NETIF_RSP_DROPPED -2
+#define XEN_NETIF_RSP_ERROR -1
+#define XEN_NETIF_RSP_OKAY 0
+/* No response: used for auxiliary requests (e.g., xen_netif_extra_info). */
+#define XEN_NETIF_RSP_NULL 1
#endif
^ permalink raw reply related
* usage setkey delete option for ipsec
From: Jack Daniel @ 2011-02-08 10:14 UTC (permalink / raw)
To: netdev
Hi,
My intention is to delete SADB entries one by one using the setkey
delete option.
I have the follwoing entry in my setkey.conf
delete fec2::1[55001] fec2:0:f675:803a::4[55002] esp 4101;
But when running $setkey -f setkey.conf
it throws scuh an error ...
line 1: syntax error at [[]
parse failed, line 1.
If I remove the port values [55001] and [55002] then setkey says
"The result of line 1: No such process."
My $setkey -D
output is shown below
setkey -D
fec2:0:888b:5e45::6 fec2::1
esp mode=transport spi=4106(0x0000100a) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=7 pid=18112 refcnt=0
fec2:0:888b:5e44::5 fec2::1
esp mode=transport spi=4104(0x00001008) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=6 pid=18112 refcnt=0
fec2:0:888b:5e43::4 fec2::1
esp mode=transport spi=4102(0x00001006) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=5 pid=18112 refcnt=0
fec2:0:888b:5e42::3 fec2::1
esp mode=transport spi=4100(0x00001004) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=4 pid=18112 refcnt=0
fec2::1 fec2:0:888b:5e42::3
esp mode=transport spi=4099(0x00001003) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=3 pid=18112 refcnt=0
fec2::1 fec2:0:888b:5e43::4
esp mode=transport spi=4101(0x00001005) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=2 pid=18112 refcnt=0
fec2::1 fec2:0:888b:5e44::5
esp mode=transport spi=4103(0x00001007) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=1 pid=18112 refcnt=0
fec2::1 fec2:0:888b:5e45::6
esp mode=transport spi=4105(0x00001009) reqid=0(0x00000000)
A: hmac-sha1 41482053 4120636f 6e666967 75726174 696f6e21
seq=0x00000000 replay=0 flags=0x00000000 state=mature
created: Feb 7 15:33:45 2011 current: Feb 7 17:09:22 2011
diff: 5737(s) hard: 0(s) soft: 0(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=0 pid=18112 refcnt=0
Could someone point out what I might be doing wrong or the right way
to pass the "delete" option an ipv6 address along with the port in
square brackets.
Thanks and regards,
Jack.
^ permalink raw reply
* Re: [PATCH] tcp: undo_retrans counter fixes
From: Ilpo Järvinen @ 2011-02-08 9:54 UTC (permalink / raw)
To: Yuchung Cheng; +Cc: David Miller, Netdev
In-Reply-To: <AANLkTikh4YXzJGgTWsTjpz6BczrfSGYD_qSg3WWmKrs9@mail.gmail.com>
[-- Attachment #1: Type: TEXT/PLAIN, Size: 4198 bytes --]
On Mon, 7 Feb 2011, Yuchung Cheng wrote:
> On Mon, Feb 7, 2011 at 3:36 PM, Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> wrote:
> >
> > On Mon, 7 Feb 2011, David Miller wrote:
> >
> > > From: Yuchung Cheng <ycheng@google.com>
> > > Date: Mon, 7 Feb 2011 14:57:04 -0800
> > >
> > > > Fix a bug that undo_retrans is incorrectly decremented when undo_marker is
> > > > not set or undo_retrans is already 0. This happens when sender receives
> > > > more DSACK ACKs than packets retransmitted during the current
> > > > undo phase. This may also happen when sender receives DSACK after
> > > > the undo operation is completed or cancelled.
> > > >
> > > > Fix another bug that undo_retrans is incorrectly incremented when
> > > > sender retransmits an skb and tcp_skb_pcount(skb) > 1 (TSO). This case
> > > > is rare but not impossible.
> > > >
> > > > Signed-off-by: Yuchung Cheng <ycheng@google.com>
> >
> > Neither is harmful to "fix" but I think they're partially also checking
> > for things which shouldn't cause problems... E.g., undo_retrans is only
> > used after checking undo_marker's validity first so I don't think
> > undo_marker check is exactly necessary there (but on the other hand it
> > does no harm)...
>
> logically we should check the validity of undo_marker/undo_retrans
> before we use them? The current code has no problem if
> tcp_fastretrans_alert() always call tcp_try_undo_* functions whenever
> undo_marker != 0 and undo_retrans == 0. I don't think that's always
> true.
We certainly should be letting the undo_retrans to underflow that in this
your patch has merit (the added tp->undo_retrans check).
However, the only users are:
static inline int tcp_may_undo(struct tcp_sock *tp)
{
return tp->undo_marker && (!tp->undo_retrans ...)
and:
static void tcp_try_undo_dsack(struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
if (tp->undo_marker && !tp->undo_retrans) {
...which check that undo_retrans is valid.
> > The tcp_retransmit_skb problem I don't understand at all as we should be
> > fragmenting or resetting pcount to 1 (the latter is true only if all
> > bugfixes were included to the kernel where >1 pcount for a rexmitted skb
> > was seen). If pcount is indeed >1 we might have other issues too somewhere
>
> We found that sometimes pcount > 1 on real servers. This change keeps
> the retrans_out/undo_retrans counters consistent.
There's still some bug then I guess... It might be related to the issues
seen by those other guys who were complaining about small segments with
>1 pcount breaking their hardware (few months ago). For the record, the
last fix is from 2.6.31 or so.
Like I said, I don't oppose this change anyway:
> > but I fail to remember immediately what they would be. That change is not
> > bad though since using +/-1 is something we should be getting rid of
> > anyway and on long term it would be nice to make tcp_retransmit_skb to be
> > able to take advantage of TSO anyway whenever possible.
...it certainly won't hurt to be on the safe side here if/when something
else is wrong.
> > I also noticed that the undo_retrans code in sacktag side is still doing
> > undo_retrans-- ops which could certainly cause real miscounts, though
> > it is extremely unlikely due to the fact that DSACK should be sent
> > immediately for a single segment at a time (so the sender would need to
> > split+recollapse in between).
>
> I have the same doubt but our servers never hit this condition (pcount
> 1). So I keep this part intact.
I could think of some scenario you cannot even reproduce in a large scale
tests, unlikely indeed :-). ...Or too stable connectivity on the sender
side. But I've changed my mind... the -1 operation is the correct one as
we could otherwise overestimate due to pcount=1->2 split after the
retransmission that triggered the DSACK (now that I remember this, I think
I've once thought this line through already earlier... I'll try to write a
comment one day there).
For -rc/next purposes I don't see any big enough reasons to withhold:
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
...but if you want this to stables too I don't think it's minimal w.r.t.
undo_marker check.
--
i.
^ permalink raw reply
* [net-2.6 6/7] ixgbe: cleanup variable initialization
From: Jeff Kirsher @ 2011-02-08 9:44 UTC (permalink / raw)
To: davem; +Cc: Don Skidmore, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297158279-20277-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Don Skidmore <donald.c.skidmore@intel.com>
The ixgbe_fcoe_ddp_get function wasn't initializing one of its variables
and this was producing compiler warnings. This patch cleans that up.
Signed-off-by: Don Skidmore <donald.c.skidmore@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_fcoe.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_fcoe.c b/drivers/net/ixgbe/ixgbe_fcoe.c
index 6342d48..8753980 100644
--- a/drivers/net/ixgbe/ixgbe_fcoe.c
+++ b/drivers/net/ixgbe/ixgbe_fcoe.c
@@ -165,7 +165,7 @@ int ixgbe_fcoe_ddp_get(struct net_device *netdev, u16 xid,
unsigned int thisoff = 0;
unsigned int thislen = 0;
u32 fcbuff, fcdmarw, fcfltrw;
- dma_addr_t addr;
+ dma_addr_t addr = 0;
if (!netdev || !sgl)
return 0;
--
1.7.4
^ permalink raw reply related
* [net-2.6 7/7] ixgbe: update version string
From: Jeff Kirsher @ 2011-02-08 9:44 UTC (permalink / raw)
To: davem; +Cc: Don Skidmore, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297158279-20277-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Don Skidmore <donald.c.skidmore@intel.com>
This will synchronize the version string with that of the latest source
forge driver which shares its functionality.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_main.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index d6c21a3..fbae703 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -52,7 +52,7 @@ char ixgbe_driver_name[] = "ixgbe";
static const char ixgbe_driver_string[] =
"Intel(R) 10 Gigabit PCI Express Network Driver";
-#define DRV_VERSION "3.0.12-k2"
+#define DRV_VERSION "3.2.9-k2"
const char ixgbe_driver_version[] = DRV_VERSION;
static char ixgbe_copyright[] = "Copyright (c) 1999-2010 Intel Corporation.";
--
1.7.4
^ permalink raw reply related
* [net-2.6 5/7] ixgbe: limit VF access to network traffic
From: Jeff Kirsher @ 2011-02-08 9:44 UTC (permalink / raw)
To: davem; +Cc: Alexander Duyck, netdev, gospo, bphilips, Jeff Kirsher
In-Reply-To: <1297158279-20277-1-git-send-email-jeffrey.t.kirsher@intel.com>
From: Alexander Duyck <alexander.h.duyck@intel.com>
This change fixes VM pool allocation issues based on MAC address filtering,
as well as limits the scope of VF access to promiscuous mode.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Acked-by: Greg Rose <gregory.v.rose@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ixgbe/ixgbe_common.c | 3 +++
drivers/net/ixgbe/ixgbe_sriov.c | 2 --
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_common.c b/drivers/net/ixgbe/ixgbe_common.c
index d5ede2d..ebbda7d 100644
--- a/drivers/net/ixgbe/ixgbe_common.c
+++ b/drivers/net/ixgbe/ixgbe_common.c
@@ -1370,6 +1370,9 @@ s32 ixgbe_init_rx_addrs_generic(struct ixgbe_hw *hw)
hw_dbg(hw, " New MAC Addr =%pM\n", hw->mac.addr);
hw->mac.ops.set_rar(hw, 0, hw->mac.addr, 0, IXGBE_RAH_AV);
+
+ /* clear VMDq pool/queue selection for RAR 0 */
+ hw->mac.ops.clear_vmdq(hw, 0, IXGBE_CLEAR_VMDQ_ALL);
}
hw->addr_ctrl.overflow_promisc = 0;
diff --git a/drivers/net/ixgbe/ixgbe_sriov.c b/drivers/net/ixgbe/ixgbe_sriov.c
index 47b1573..187b3a1 100644
--- a/drivers/net/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ixgbe/ixgbe_sriov.c
@@ -110,12 +110,10 @@ static int ixgbe_set_vf_vlan(struct ixgbe_adapter *adapter, int add, int vid,
return adapter->hw.mac.ops.set_vfta(&adapter->hw, vid, vf, (bool)add);
}
-
static void ixgbe_set_vmolr(struct ixgbe_hw *hw, u32 vf, bool aupe)
{
u32 vmolr = IXGBE_READ_REG(hw, IXGBE_VMOLR(vf));
vmolr |= (IXGBE_VMOLR_ROMPE |
- IXGBE_VMOLR_ROPE |
IXGBE_VMOLR_BAM);
if (aupe)
vmolr |= IXGBE_VMOLR_AUPE;
--
1.7.4
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox