* [patch net-next 09/16] qlcnic: guard __vlan_find_dev_deep() by rcu_read_lock
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
rcu_read_lock was missing here
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index 212c121..9eadf17 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -4425,12 +4425,14 @@ qlcnic_restore_indev_addr(struct net_device *netdev, unsigned long event)
qlcnic_config_indev_addr(adapter, netdev, event);
+ rcu_read_lock();
for_each_set_bit(vid, adapter->vlans, VLAN_N_VID) {
dev = __vlan_find_dev_deep(netdev, vid);
if (!dev)
continue;
qlcnic_config_indev_addr(adapter, dev, event);
}
+ rcu_read_unlock();
}
static int qlcnic_netdev_event(struct notifier_block *this,
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 10/16] qeth: ensure that __vlan_find_dev_deep() is called with rcu_read_lock
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Also benefit from rcu_read_lock held and use __in_dev_get_rcu() in ipv4 case.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/s390/net/qeth_l3_main.c | 21 +++++++++++----------
1 file changed, 11 insertions(+), 10 deletions(-)
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index c5f03fa..6b3988b 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1638,6 +1638,7 @@ static void qeth_l3_add_mc(struct qeth_card *card, struct in_device *in4_dev)
}
}
+/* called with rcu_read_lock */
static void qeth_l3_add_vlan_mc(struct qeth_card *card)
{
struct in_device *in_dev;
@@ -1650,19 +1651,14 @@ static void qeth_l3_add_vlan_mc(struct qeth_card *card)
for_each_set_bit(vid, card->active_vlans, VLAN_N_VID) {
struct net_device *netdev;
- rcu_read_lock();
netdev = __vlan_find_dev_deep(card->dev, vid);
- rcu_read_unlock();
if (netdev == NULL ||
!(netdev->flags & IFF_UP))
continue;
- in_dev = in_dev_get(netdev);
+ in_dev = __in_dev_get_rcu(netdev);
if (!in_dev)
continue;
- rcu_read_lock();
qeth_l3_add_mc(card, in_dev);
- rcu_read_unlock();
- in_dev_put(in_dev);
}
}
@@ -1671,14 +1667,14 @@ static void qeth_l3_add_multicast_ipv4(struct qeth_card *card)
struct in_device *in4_dev;
QETH_CARD_TEXT(card, 4, "chkmcv4");
- in4_dev = in_dev_get(card->dev);
- if (in4_dev == NULL)
- return;
rcu_read_lock();
+ in4_dev = __in_dev_get_rcu(card->dev);
+ if (in4_dev == NULL)
+ goto unlock;
qeth_l3_add_mc(card, in4_dev);
qeth_l3_add_vlan_mc(card);
+unlock:
rcu_read_unlock();
- in_dev_put(in4_dev);
}
#ifdef CONFIG_QETH_IPV6
@@ -1703,6 +1699,7 @@ static void qeth_l3_add_mc6(struct qeth_card *card, struct inet6_dev *in6_dev)
}
}
+/* called with rcu_read_lock */
static void qeth_l3_add_vlan_mc6(struct qeth_card *card)
{
struct inet6_dev *in_dev;
@@ -1739,10 +1736,12 @@ static void qeth_l3_add_multicast_ipv6(struct qeth_card *card)
in6_dev = in6_dev_get(card->dev);
if (in6_dev == NULL)
return;
+ rcu_read_lock();
read_lock_bh(&in6_dev->lock);
qeth_l3_add_mc6(card, in6_dev);
qeth_l3_add_vlan_mc6(card);
read_unlock_bh(&in6_dev->lock);
+ rcu_read_unlock();
in6_dev_put(in6_dev);
}
#endif /* CONFIG_QETH_IPV6 */
@@ -1811,8 +1810,10 @@ static void qeth_l3_free_vlan_addresses6(struct qeth_card *card,
static void qeth_l3_free_vlan_addresses(struct qeth_card *card,
unsigned short vid)
{
+ rcu_read_lock();
qeth_l3_free_vlan_addresses4(card, vid);
qeth_l3_free_vlan_addresses6(card, vid);
+ rcu_read_unlock();
}
static int qeth_l3_vlan_rx_add_vid(struct net_device *dev, unsigned short vid)
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 11/16] vlan: remove usage of dev->master in __vlan_find_dev_deep()
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Also, since all users call __vlan_find_dev_deep() with rcu_read_lock,
make no possibility to call this with rtnl mutex held only.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
net/8021q/vlan_core.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/net/8021q/vlan_core.c b/net/8021q/vlan_core.c
index 8ca533c..f9518ad 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -66,21 +66,25 @@ bool vlan_do_receive(struct sk_buff **skbp, bool last_handler)
return true;
}
-/* Must be invoked with rcu_read_lock or with RTNL. */
-struct net_device *__vlan_find_dev_deep(struct net_device *real_dev,
+/* Must be invoked with rcu_read_lock. */
+struct net_device *__vlan_find_dev_deep(struct net_device *dev,
u16 vlan_id)
{
- struct vlan_info *vlan_info = rcu_dereference_rtnl(real_dev->vlan_info);
+ struct vlan_info *vlan_info = rcu_dereference(dev->vlan_info);
if (vlan_info) {
return vlan_group_get_device(&vlan_info->grp, vlan_id);
} else {
/*
- * Bonding slaves do not have grp assigned to themselves.
- * Grp is assigned to bonding master instead.
+ * Lower devices of unique uppers (bonding, team) do not have
+ * grp assigned to themselves. Grp is assigned to upper device
+ * instead.
*/
- if (netif_is_bond_slave(real_dev))
- return __vlan_find_dev_deep(real_dev->master, vlan_id);
+ struct net_device *upper_dev;
+
+ upper_dev = netdev_unique_upper_dev_get_rcu(dev);
+ if (upper_dev)
+ return __vlan_find_dev_deep(upper_dev, vlan_id);
}
return NULL;
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 12/16] nes: remove usage of dev->master
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/infiniband/hw/nes/nes.c | 8 +++++---
drivers/infiniband/hw/nes/nes_cm.c | 2 +-
2 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c
index 7140199..e58e9af 100644
--- a/drivers/infiniband/hw/nes/nes.c
+++ b/drivers/infiniband/hw/nes/nes.c
@@ -140,6 +140,7 @@ static int nes_inetaddr_event(struct notifier_block *notifier,
struct net_device *event_netdev = ifa->ifa_dev->dev;
struct nes_device *nesdev;
struct net_device *netdev;
+ struct net_device *upper_dev;
struct nes_vnic *nesvnic;
unsigned int is_bonded;
@@ -150,8 +151,9 @@ static int nes_inetaddr_event(struct notifier_block *notifier,
nesdev, nesdev->netdev[0]->name);
netdev = nesdev->netdev[0];
nesvnic = netdev_priv(netdev);
+ upper_dev = netdev_unique_upper_dev_get(netdev);
is_bonded = netif_is_bond_slave(netdev) &&
- (netdev->master == event_netdev);
+ (upper_dev == event_netdev);
if ((netdev == event_netdev) || is_bonded) {
if (nesvnic->rdma_enabled == 0) {
nes_debug(NES_DBG_NETDEV, "Returning without processing event for %s since"
@@ -184,9 +186,9 @@ static int nes_inetaddr_event(struct notifier_block *notifier,
/* fall through */
case NETDEV_CHANGEADDR:
/* Add the address to the IP table */
- if (netdev->master)
+ if (upper_dev)
nesvnic->local_ipaddr =
- ((struct in_device *)netdev->master->ip_ptr)->ifa_list->ifa_address;
+ ((struct in_device *)upper_dev->ip_ptr)->ifa_list->ifa_address;
else
nesvnic->local_ipaddr = ifa->ifa_address;
diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c
index 020e95c..15e6340 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -1352,7 +1352,7 @@ static int nes_addr_resolve_neigh(struct nes_vnic *nesvnic, u32 dst_ip, int arpi
}
if (netif_is_bond_slave(nesvnic->netdev))
- netdev = nesvnic->netdev->master;
+ netdev = netdev_unique_upper_dev_get(nesvnic->netdev);
else
netdev = nesvnic->netdev;
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 13/16] bonding: remove usage of dev->master
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Benefit from new upper dev list and free bonding from dev->master usage.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/net/bonding/bond_3ad.c | 30 ++++++-------
drivers/net/bonding/bond_alb.c | 6 +--
drivers/net/bonding/bond_main.c | 94 +++++++++++++++++++++++----------------
drivers/net/bonding/bonding.h | 14 +++---
net/core/rtnetlink.c | 1 +
5 files changed, 81 insertions(+), 64 deletions(-)
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index a030e63..84fabd6 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -1127,7 +1127,7 @@ static void ad_rx_machine(struct lacpdu *lacpdu, struct port *port)
// INFO_RECEIVED_LOOPBACK_FRAMES
pr_err("%s: An illegal loopback occurred on adapter (%s).\n"
"Check the configuration to verify that all adapters are connected to 802.3ad compliant switch ports\n",
- port->slave->dev->master->name, port->slave->dev->name);
+ port->slave->bond->dev->name, port->slave->dev->name);
return;
}
__update_selected(lacpdu, port);
@@ -1306,7 +1306,7 @@ static void ad_port_selection_logic(struct port *port)
}
if (!curr_port) { // meaning: the port was related to an aggregator but was not on the aggregator port list
pr_warning("%s: Warning: Port %d (on %s) was related to aggregator %d but was not on its port list\n",
- port->slave->dev->master->name,
+ port->slave->bond->dev->name,
port->actor_port_number,
port->slave->dev->name,
port->aggregator->aggregator_identifier);
@@ -1386,7 +1386,7 @@ static void ad_port_selection_logic(struct port *port)
port->aggregator->aggregator_identifier);
} else {
pr_err("%s: Port %d (on %s) did not find a suitable aggregator\n",
- port->slave->dev->master->name,
+ port->slave->bond->dev->name,
port->actor_port_number, port->slave->dev->name);
}
}
@@ -1463,7 +1463,7 @@ static struct aggregator *ad_agg_selection_test(struct aggregator *best,
default:
pr_warning("%s: Impossible agg select mode %d\n",
- curr->slave->dev->master->name,
+ curr->slave->bond->dev->name,
__get_agg_selection_mode(curr->lag_ports));
break;
}
@@ -1571,7 +1571,7 @@ static void ad_agg_selection_logic(struct aggregator *agg)
// check if any partner replys
if (best->is_individual) {
pr_warning("%s: Warning: No 802.3ad response from the link partner for any adapters in the bond\n",
- best->slave ? best->slave->dev->master->name : "NULL");
+ best->slave ? best->slave->bond->dev->name : "NULL");
}
best->is_active = 1;
@@ -1898,7 +1898,7 @@ int bond_3ad_bind_slave(struct slave *slave)
if (bond == NULL) {
pr_err("%s: The slave %s is not attached to its bond\n",
- slave->dev->master->name, slave->dev->name);
+ slave->bond->dev->name, slave->dev->name);
return -1;
}
@@ -1973,7 +1973,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
// if slave is null, the whole port is not initialized
if (!port->slave) {
pr_warning("Warning: %s: Trying to unbind an uninitialized port on %s\n",
- slave->dev->master->name, slave->dev->name);
+ slave->bond->dev->name, slave->dev->name);
return;
}
@@ -2009,7 +2009,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
if ((new_aggregator->lag_ports == port) && new_aggregator->is_active) {
pr_info("%s: Removing an active aggregator\n",
- aggregator->slave->dev->master->name);
+ aggregator->slave->bond->dev->name);
// select new active aggregator
select_new_active_agg = 1;
}
@@ -2040,7 +2040,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
ad_agg_selection_logic(__get_first_agg(port));
} else {
pr_warning("%s: Warning: unbinding aggregator, and could not find a new aggregator for its ports\n",
- slave->dev->master->name);
+ slave->bond->dev->name);
}
} else { // in case that the only port related to this aggregator is the one we want to remove
select_new_active_agg = aggregator->is_active;
@@ -2048,7 +2048,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
ad_clear_agg(aggregator);
if (select_new_active_agg) {
pr_info("%s: Removing an active aggregator\n",
- slave->dev->master->name);
+ slave->bond->dev->name);
// select new active aggregator
ad_agg_selection_logic(__get_first_agg(port));
}
@@ -2076,7 +2076,7 @@ void bond_3ad_unbind_slave(struct slave *slave)
ad_clear_agg(temp_aggregator);
if (select_new_active_agg) {
pr_info("%s: Removing an active aggregator\n",
- slave->dev->master->name);
+ slave->bond->dev->name);
// select new active aggregator
ad_agg_selection_logic(__get_first_agg(port));
}
@@ -2184,7 +2184,7 @@ static int bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave, u1
if (!port->slave) {
pr_warning("%s: Warning: port of slave %s is uninitialized\n",
- slave->dev->name, slave->dev->master->name);
+ slave->dev->name, slave->bond->dev->name);
return ret;
}
@@ -2240,7 +2240,7 @@ void bond_3ad_adapter_speed_changed(struct slave *slave)
// if slave is null, the whole port is not initialized
if (!port->slave) {
pr_warning("Warning: %s: speed changed for uninitialized port on %s\n",
- slave->dev->master->name, slave->dev->name);
+ slave->bond->dev->name, slave->dev->name);
return;
}
@@ -2268,7 +2268,7 @@ void bond_3ad_adapter_duplex_changed(struct slave *slave)
// if slave is null, the whole port is not initialized
if (!port->slave) {
pr_warning("%s: Warning: duplex changed for uninitialized port on %s\n",
- slave->dev->master->name, slave->dev->name);
+ slave->bond->dev->name, slave->dev->name);
return;
}
@@ -2297,7 +2297,7 @@ void bond_3ad_handle_link_change(struct slave *slave, char link)
// if slave is null, the whole port is not initialized
if (!port->slave) {
pr_warning("Warning: %s: link status changed for uninitialized port on %s\n",
- slave->dev->master->name, slave->dev->name);
+ slave->bond->dev->name, slave->dev->name);
return;
}
diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index e15cc11..6dc3625 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -491,7 +491,7 @@ static void rlb_update_client(struct rlb_client_info *client_info)
client_info->mac_dst);
if (!skb) {
pr_err("%s: Error: failed to create an ARP packet\n",
- client_info->slave->dev->master->name);
+ client_info->slave->bond->dev->name);
continue;
}
@@ -501,7 +501,7 @@ static void rlb_update_client(struct rlb_client_info *client_info)
skb = vlan_put_tag(skb, client_info->vlan_id);
if (!skb) {
pr_err("%s: Error: failed to insert VLAN tag\n",
- client_info->slave->dev->master->name);
+ client_info->slave->bond->dev->name);
continue;
}
}
@@ -914,7 +914,7 @@ static int alb_set_slave_mac_addr(struct slave *slave, u8 addr[])
if (dev_set_mac_address(dev, &s_addr)) {
pr_err("%s: Error: dev_set_mac_address of dev %s failed!\n"
"ALB mode requires that the base driver support setting the hw address also when the network device's interface is open\n",
- dev->master->name, dev->name);
+ slave->bond->dev->name, dev->name);
return -EOPNOTSUPP;
}
return 0;
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 6fae5f3..de75541 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -752,11 +752,9 @@ static void __bond_resend_igmp_join_requests(struct net_device *dev)
{
struct in_device *in_dev;
- rcu_read_lock();
in_dev = __in_dev_get_rcu(dev);
if (in_dev)
ip_mc_rejoin_groups(in_dev);
- rcu_read_unlock();
}
/*
@@ -766,9 +764,10 @@ static void __bond_resend_igmp_join_requests(struct net_device *dev)
*/
static void bond_resend_igmp_join_requests(struct bonding *bond)
{
- struct net_device *bond_dev, *vlan_dev, *master_dev;
+ struct net_device *bond_dev, *vlan_dev, *upper_dev;
struct vlan_entry *vlan;
+ rcu_read_lock();
read_lock(&bond->lock);
bond_dev = bond->dev;
@@ -780,18 +779,14 @@ static void bond_resend_igmp_join_requests(struct bonding *bond)
* if bond is enslaved to a bridge,
* then rejoin all groups on its master
*/
- master_dev = bond_dev->master;
- if (master_dev)
- if ((master_dev->priv_flags & IFF_EBRIDGE)
- && (bond_dev->priv_flags & IFF_BRIDGE_PORT))
- __bond_resend_igmp_join_requests(master_dev);
+ upper_dev = netdev_unique_upper_dev_get_rcu(bond_dev);
+ if (upper_dev && upper_dev->priv_flags & IFF_EBRIDGE)
+ __bond_resend_igmp_join_requests(upper_dev);
/* rejoin all groups on vlan devices */
list_for_each_entry(vlan, &bond->vlan_list, vlan_list) {
- rcu_read_lock();
vlan_dev = __vlan_find_dev_deep(bond_dev,
vlan->vlan_id);
- rcu_read_unlock();
if (vlan_dev)
__bond_resend_igmp_join_requests(vlan_dev);
}
@@ -800,13 +795,16 @@ static void bond_resend_igmp_join_requests(struct bonding *bond)
queue_delayed_work(bond->wq, &bond->mcast_work, HZ/5);
read_unlock(&bond->lock);
+ rcu_read_unlock();
}
static void bond_resend_igmp_join_requests_delayed(struct work_struct *work)
{
struct bonding *bond = container_of(work, struct bonding,
mcast_work.work);
+ rcu_read_lock();
bond_resend_igmp_join_requests(bond);
+ rcu_read_unlock();
}
/*
@@ -1494,6 +1492,27 @@ static rx_handler_result_t bond_handle_frame(struct sk_buff **pskb)
return ret;
}
+static int bond_unique_upper_dev_link(struct net_device *bond_dev,
+ struct net_device *slave_dev)
+{
+ int err;
+
+ err = netdev_unique_upper_dev_link(slave_dev, bond_dev);
+ if (err)
+ return err;
+ slave_dev->flags |= IFF_SLAVE;
+ rtmsg_ifinfo(RTM_NEWLINK, slave_dev, IFF_SLAVE);
+ return 0;
+}
+
+static void bond_upper_dev_unlink(struct net_device *bond_dev,
+ struct net_device *slave_dev)
+{
+ netdev_upper_dev_unlink(slave_dev, bond_dev);
+ slave_dev->flags &= ~IFF_SLAVE;
+ rtmsg_ifinfo(RTM_NEWLINK, slave_dev, IFF_SLAVE);
+}
+
/* enslave device <slave> to bond device <master> */
int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
{
@@ -1655,9 +1674,9 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
}
}
- res = netdev_set_bond_master(slave_dev, bond_dev);
+ res = bond_unique_upper_dev_link(bond_dev, slave_dev);
if (res) {
- pr_debug("Error %d calling netdev_set_bond_master\n", res);
+ pr_debug("Error %d calling bond_unique_upper_dev_link\n", res);
goto err_restore_mac;
}
@@ -1891,7 +1910,7 @@ err_close:
dev_close(slave_dev);
err_unset_master:
- netdev_set_bond_master(slave_dev, NULL);
+ bond_upper_dev_unlink(bond_dev, slave_dev);
err_restore_mac:
if (!bond->params.fail_over_mac) {
@@ -1936,7 +1955,7 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
/* slave is not a slave or master is not master of this slave */
if (!(slave_dev->flags & IFF_SLAVE) ||
- (slave_dev->master != bond_dev)) {
+ !netdev_has_upper_dev(slave_dev, bond_dev)) {
pr_err("%s: Error: cannot release %s.\n",
bond_dev->name, slave_dev->name);
return -EINVAL;
@@ -2080,7 +2099,7 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
netif_addr_unlock_bh(bond_dev);
}
- netdev_set_bond_master(slave_dev, NULL);
+ bond_upper_dev_unlink(bond_dev, slave_dev);
slave_disable_netpoll(slave);
@@ -2195,7 +2214,7 @@ static int bond_release_all(struct net_device *bond_dev)
netif_addr_unlock_bh(bond_dev);
}
- netdev_set_bond_master(slave_dev, NULL);
+ bond_upper_dev_unlink(bond_dev, slave_dev);
slave_disable_netpoll(slave);
@@ -2259,8 +2278,9 @@ static int bond_ioctl_change_active(struct net_device *bond_dev, struct net_devi
if (!USES_PRIMARY(bond->params.mode))
return -EINVAL;
- /* Verify that master_dev is indeed the master of slave_dev */
- if (!(slave_dev->flags & IFF_SLAVE) || (slave_dev->master != bond_dev))
+ /* Verify that bond_dev is indeed the master of slave_dev */
+ if (!(slave_dev->flags & IFF_SLAVE) ||
+ !netdev_has_upper_dev(slave_dev, bond_dev))
return -EINVAL;
read_lock(&bond->lock);
@@ -3249,36 +3269,32 @@ static int bond_master_netdev_event(unsigned long event,
static int bond_slave_netdev_event(unsigned long event,
struct net_device *slave_dev)
{
- struct net_device *bond_dev = slave_dev->master;
- struct bonding *bond = netdev_priv(bond_dev);
- struct slave *slave = NULL;
+ struct slave *slave = bond_slave_get_rtnl(slave_dev);
+ struct bonding *bond = slave->bond;
+ struct net_device *bond_dev = slave->bond->dev;
+ u32 old_speed;
+ u8 old_duplex;
switch (event) {
case NETDEV_UNREGISTER:
- if (bond_dev) {
- if (bond->setup_by_slave)
- bond_release_and_destroy(bond_dev, slave_dev);
- else
- bond_release(bond_dev, slave_dev);
- }
+ if (bond->setup_by_slave)
+ bond_release_and_destroy(bond_dev, slave_dev);
+ else
+ bond_release(bond_dev, slave_dev);
break;
case NETDEV_UP:
case NETDEV_CHANGE:
- slave = bond_get_slave_by_dev(bond, slave_dev);
- if (slave) {
- u32 old_speed = slave->speed;
- u8 old_duplex = slave->duplex;
+ old_speed = slave->speed;
+ old_duplex = slave->duplex;
- bond_update_speed_duplex(slave);
+ bond_update_speed_duplex(slave);
- if (bond->params.mode == BOND_MODE_8023AD) {
- if (old_speed != slave->speed)
- bond_3ad_adapter_speed_changed(slave);
- if (old_duplex != slave->duplex)
- bond_3ad_adapter_duplex_changed(slave);
- }
+ if (bond->params.mode == BOND_MODE_8023AD) {
+ if (old_speed != slave->speed)
+ bond_3ad_adapter_speed_changed(slave);
+ if (old_duplex != slave->duplex)
+ bond_3ad_adapter_duplex_changed(slave);
}
-
break;
case NETDEV_DOWN:
/*
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index f8af2fc..c9cd1c0 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -257,6 +257,9 @@ static inline bool bond_vlan_used(struct bonding *bond)
#define bond_slave_get_rcu(dev) \
((struct slave *) rcu_dereference(dev->rx_handler_data))
+#define bond_slave_get_rtnl(dev) \
+ ((struct slave *) rtnl_dereference(dev->rx_handler_data))
+
/**
* Returns NULL if the net_device does not belong to any of the bond's slaves
*
@@ -279,11 +282,9 @@ static inline struct slave *bond_get_slave_by_dev(struct bonding *bond,
static inline struct bonding *bond_get_bond_by_slave(struct slave *slave)
{
- if (!slave || !slave->dev->master) {
+ if (!slave || !slave->bond)
return NULL;
- }
-
- return netdev_priv(slave->dev->master);
+ return slave->bond;
}
static inline bool bond_is_lb(const struct bonding *bond)
@@ -359,10 +360,9 @@ static inline void bond_netpoll_send_skb(const struct slave *slave,
static inline void bond_set_slave_inactive_flags(struct slave *slave)
{
- struct bonding *bond = netdev_priv(slave->dev->master);
- if (!bond_is_lb(bond))
+ if (!bond_is_lb(slave->bond))
bond_set_backup_slave(slave);
- if (!bond->params.all_slaves_active)
+ if (!slave->bond->params.all_slaves_active)
slave->inactive = 1;
}
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index c341eb7..df16052 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1989,6 +1989,7 @@ errout:
if (err < 0)
rtnl_set_sk_err(net, RTNLGRP_LINK, err);
}
+EXPORT_SYMBOL(rtmsg_ifinfo);
static int nlmsg_populate_fdb_fill(struct sk_buff *skb,
struct net_device *dev,
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 14/16] net: remove no longer used netdev_set_bond_master() and netdev_set_master()
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
include/linux/netdevice.h | 6 +----
net/core/dev.c | 63 ---------------------------------------------
2 files changed, 1 insertion(+), 68 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index e7a07f8..757f627 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -886,8 +886,7 @@ struct netdev_fcoe_hbainfo {
* flow_id is a flow ID to be passed to rps_may_expire_flow() later.
* Return the filter ID on success, or a negative error code.
*
- * Slave management functions (for bridge, bonding, etc). User should
- * call netdev_set_master() to set dev->master properly.
+ * Slave management functions (for bridge, bonding, etc).
* int (*ndo_add_slave)(struct net_device *dev, struct net_device *slave_dev);
* Called to make another netdev an underling.
*
@@ -2625,9 +2624,6 @@ extern int netdev_unique_upper_dev_link(struct net_device *dev,
struct net_device *upper_dev);
extern void netdev_upper_dev_unlink(struct net_device *dev,
struct net_device *upper_dev);
-extern int netdev_set_master(struct net_device *dev, struct net_device *master);
-extern int netdev_set_bond_master(struct net_device *dev,
- struct net_device *master);
extern int skb_checksum_help(struct sk_buff *skb);
extern struct sk_buff *skb_gso_segment(struct sk_buff *skb,
netdev_features_t features);
diff --git a/net/core/dev.c b/net/core/dev.c
index 68db1ac..c0f9adb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4648,69 +4648,6 @@ void netdev_upper_dev_unlink(struct net_device *dev,
}
EXPORT_SYMBOL(netdev_upper_dev_unlink);
-/**
- * netdev_set_master - set up master pointer
- * @slave: slave device
- * @master: new master device
- *
- * Changes the master device of the slave. Pass %NULL to break the
- * bonding. The caller must hold the RTNL semaphore. On a failure
- * a negative errno code is returned. On success the reference counts
- * are adjusted and the function returns zero.
- */
-int netdev_set_master(struct net_device *slave, struct net_device *master)
-{
- struct net_device *old = slave->master;
- int err;
-
- ASSERT_RTNL();
-
- if (master) {
- if (old)
- return -EBUSY;
- err = netdev_unique_upper_dev_link(slave, master);
- if (err)
- return err;
- }
-
- slave->master = master;
-
- if (old)
- netdev_upper_dev_unlink(slave, master);
-
- return 0;
-}
-EXPORT_SYMBOL(netdev_set_master);
-
-/**
- * netdev_set_bond_master - set up bonding master/slave pair
- * @slave: slave device
- * @master: new master device
- *
- * Changes the master device of the slave. Pass %NULL to break the
- * bonding. The caller must hold the RTNL semaphore. On a failure
- * a negative errno code is returned. On success %RTM_NEWLINK is sent
- * to the routing socket and the function returns zero.
- */
-int netdev_set_bond_master(struct net_device *slave, struct net_device *master)
-{
- int err;
-
- ASSERT_RTNL();
-
- err = netdev_set_master(slave, master);
- if (err)
- return err;
- if (master)
- slave->flags |= IFF_SLAVE;
- else
- slave->flags &= ~IFF_SLAVE;
-
- rtmsg_ifinfo(RTM_NEWLINK, slave, IFF_SLAVE);
- return 0;
-}
-EXPORT_SYMBOL(netdev_set_bond_master);
-
static void dev_change_rx_flags(struct net_device *dev, int flags)
{
const struct net_device_ops *ops = dev->netdev_ops;
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 15/16] net: remove usage of dev->master
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
net/core/dev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index c0f9adb..8977404 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5461,8 +5461,8 @@ static void rollback_registered_many(struct list_head *head)
if (dev->netdev_ops->ndo_uninit)
dev->netdev_ops->ndo_uninit(dev);
- /* Notifier chain MUST detach us from master device. */
- WARN_ON(dev->master);
+ /* Notifier chain MUST detach us all upper devices. */
+ WARN_ON(netdev_has_any_upper_dev(dev));
/* Remove entries from kobject tree */
netdev_unregister_kobject(dev);
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 08/16] cxgb3: remove usage of dev->master
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
index 2dbbcbb..28365db 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
@@ -182,14 +182,17 @@ static struct net_device *get_iff_from_mac(struct adapter *adapter,
struct net_device *dev = adapter->port[i];
if (!memcmp(dev->dev_addr, mac, ETH_ALEN)) {
+ rcu_read_lock();
if (vlan && vlan != VLAN_VID_MASK) {
- rcu_read_lock();
dev = __vlan_find_dev_deep(dev, vlan);
- rcu_read_unlock();
} else if (netif_is_bond_slave(dev)) {
- while (dev->master)
- dev = dev->master;
+ struct net_device *upper_dev;
+
+ while ((upper_dev =
+ netdev_unique_upper_dev_get_rcu(dev)))
+ dev = upper_dev;
}
+ rcu_read_unlock();
return dev;
}
}
--
1.7.10.4
^ permalink raw reply related
* [patch net-next 16/16] net: kill dev->master
From: Jiri Pirko @ 2012-08-13 15:27 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, faisal.latif, roland, sean.hefty, hal.rosenstock,
fubar, andy, divy, jitendra.kalsaria, sony.chacko, linux-driver,
kaber, ursula.braun, blaschka, linux390, shemminger, bhutchings,
therbert, xiyou.wangcong, joe, gregory.v.rose, john.r.fastabend,
linux-rdma, linux-kernel, linux-s390, bridge, fbl
In-Reply-To: <1344871635-1052-1-git-send-email-jiri@resnulli.us>
Nobody uses this now. Remove it.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
include/linux/netdevice.h | 4 ----
1 file changed, 4 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 757f627..28c5ef0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1168,10 +1168,6 @@ struct net_device {
* avoid dirtying this cache line.
*/
- struct net_device *master; /* Pointer to master device of a group,
- * which this device is member of.
- */
-
struct list_head upper_dev_list; /* List of upper devices */
/* Interface address info used in eth_type_trans() */
--
1.7.10.4
^ permalink raw reply related
* Re: [PATCH] tc-tbf.8: Add parameter range to man page.
From: Stephen Hemminger @ 2012-08-13 15:30 UTC (permalink / raw)
To: Li Wei; +Cc: netdev
In-Reply-To: <1344306632-15288-1-git-send-email-lw@cn.fujitsu.com>
On Tue, 7 Aug 2012 10:30:32 +0800
Li Wei <lw@cn.fujitsu.com> wrote:
> Signed-off-by: Li Wei <lw@cn.fujitsu.com>
> ---
> man/man8/tc-tbf.8 | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/man/man8/tc-tbf.8 b/man/man8/tc-tbf.8
> index 3abb238..7b76146 100644
> --- a/man/man8/tc-tbf.8
> +++ b/man/man8/tc-tbf.8
> @@ -76,6 +76,7 @@ latency parameter, which specifies the maximum amount of time a packet can
> sit in the TBF. The latter calculation takes into account the size of the
> bucket, the rate and possibly the peakrate (if set). These two parameters
> are mutually exclusive.
> +The range of parameter limit is [1, UINT32_MAX] bytes.
> .TP
> burst
> Also known as buffer or maxburst.
> @@ -85,6 +86,7 @@ if you want to reach your configured rate!
>
> If your buffer is too small, packets may be dropped because more tokens arrive per timer tick than fit in your bucket.
> The minimum buffer size can be calculated by dividing the rate by HZ.
> +The range of this parameter is [1, UINT32_MAX] bytes.
>
> Token usage calculations are performed using a table which by default has a resolution of 8 packets.
> This resolution can be changed by specifying the
> @@ -96,11 +98,13 @@ this. Must be an integral power of 2.
> mpu
> A zero-sized packet does not use zero bandwidth. For ethernet, no packet uses less than 64 bytes. The Minimum Packet Unit
> determines the minimal token usage (specified in bytes) for a packet. Defaults to zero.
> +The range of this parameter is [0, UINT32_MAX] bytes.
> .TP
> rate
> The speed knob. See remarks above about limits! See
> .BR tc (8)
> for units.
> +The range of this parameter is [1, UINT32_MAX] bps.
> .PP
> Furthermore, if a peakrate is desired, the following parameters are available:
>
> @@ -108,12 +112,14 @@ Furthermore, if a peakrate is desired, the following parameters are available:
> peakrate
> Maximum depletion rate of the bucket. Limited to 1mbit/s on Intel, 10mbit/s on Alpha. The peakrate does
> not need to be set, it is only necessary if perfect millisecond timescale shaping is required.
> +The range of this parameter is [1, UINT32_MAX] bps.
>
> .TP
> mtu/minburst
> Specifies the size of the peakrate bucket. For perfect accuracy, should be set to the MTU of the interface.
> If a peakrate is needed, but some burstiness is acceptable, this size can be raised. A 3000 byte minburst
> allows around 3mbit/s of peakrate, given 1000 byte packets.
> +The range of this parameter is [1, UINT32_MAX] bytes.
>
> Like the regular burstsize you can also specify a
> .B cell
Not sure about this, other qdisc don't document ranges on their man pages.
Probably better to fix the generic documentation about qdisc parameters
which already exists on 'tc-qdisc' man page.
^ permalink raw reply
* Re: [Xen-devel] [PATCH] netvm: check for page == NULL when propogating the skb->pfmemalloc flag
From: Konrad Rzeszutek Wilk @ 2012-08-13 15:41 UTC (permalink / raw)
To: David Miller, Ian Campbell
Cc: mgorman, xen-devel, netdev, linux-kernel, Ian.Campbell, linux-mm,
konrad, akpm
In-Reply-To: <20120808.155046.820543563969484712.davem@davemloft.net>
On Wed, Aug 08, 2012 at 03:50:46PM -0700, David Miller wrote:
> From: Mel Gorman <mgorman@suse.de>
> Date: Tue, 7 Aug 2012 09:55:55 +0100
>
> > Commit [c48a11c7: netvm: propagate page->pfmemalloc to skb] is responsible
> > for the following bug triggered by a xen network driver
> ...
> > The problem is that the xenfront driver is passing a NULL page to
> > __skb_fill_page_desc() which was unexpected. This patch checks that
> > there is a page before dereferencing.
> >
> > Reported-and-Tested-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> > Signed-off-by: Mel Gorman <mgorman@suse.de>
>
> That call to __skb_fill_page_desc() in xen-netfront.c looks completely bogus.
> It's the only driver passing NULL here.
It looks to be passing a valid page pointer (at least by looking
at the code) so I am not sure how it got turned in a NULL.
But let me double-check by instrumenting the driver..
>
> That whole song and dance figuring out what to do with the head
> fragment page, depending upon whether the length is greater than the
> RX_COPY_THRESHOLD, is completely unnecessary.
>
> Just use something like a call to __pskb_pull_tail(skb, len) and all
> that other crap around that area can simply be deleted.
It looks like an overkill - it does a lot more than just allocate an SKB
and a page.
Deleting of extra code would be nice - however I am not going to be able
to do that for the next two weeks sadly - as my plate if full of debugging
some other stuff.
Lets see if Ian has some time.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* [PATCH net-next 0/4] packet: Sock-diag extension for packet sockets (beginning)
From: Pavel Emelyanov @ 2012-08-13 15:47 UTC (permalink / raw)
To: Linux Netdev List, David Miller
There's a set of stuff residing on a AF_PACKET socket which is currently
write-only -- rings, copy_thresh and mclist. But we need to know these
while doing the checkpoint-restore.
As a solution for the similar problem with unix sockets, the sock-diag
engine was developed, so here's the basic implementation of the packet
sockets extension. It doesn't report rings, fanout and stats, but it can
be easily added later.
I'd like to mention, that it's not a strict requirement, that the diag is
used for getting info about sockets, it would be perfectly fine just to
fix the packet getsockopt callback to return the desired info. So, if the
diag for packet sockets is for any reason unacceptable, just let me know.
Thanks,
Pavel
^ permalink raw reply
* [PATCH 1/4] packet: Introduce net/packet/internal.h header
From: Pavel Emelyanov @ 2012-08-13 15:49 UTC (permalink / raw)
To: Linux Netdev List, David Miller
In-Reply-To: <5029219C.20305@parallels.com>
The diag module will need to access some private packet_sock data, so
move it to a header in advance. This file will be shared between the
af_packet.c and the diag.c
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
net/packet/af_packet.c | 99 +---------------------------------------------
net/packet/internal.h | 103 ++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 105 insertions(+), 97 deletions(-)
create mode 100644 net/packet/internal.h
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index ceaca7c..8a1605a 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -93,6 +93,8 @@
#include <net/inet_common.h>
#endif
+#include "internal.h"
+
/*
Assumptions:
- if device has no dev->hard_header routine, it adds and removes ll header
@@ -146,14 +148,6 @@ dev->hard_header == NULL (ll header is added by device, we cannot control it)
/* Private packet socket structures. */
-struct packet_mclist {
- struct packet_mclist *next;
- int ifindex;
- int count;
- unsigned short type;
- unsigned short alen;
- unsigned char addr[MAX_ADDR_LEN];
-};
/* identical to struct packet_mreq except it has
* a longer address field.
*/
@@ -175,63 +169,7 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u,
#define BLK_PLUS_PRIV(sz_of_priv) \
(BLK_HDR_LEN + ALIGN((sz_of_priv), V3_ALIGNMENT))
-/* kbdq - kernel block descriptor queue */
-struct tpacket_kbdq_core {
- struct pgv *pkbdq;
- unsigned int feature_req_word;
- unsigned int hdrlen;
- unsigned char reset_pending_on_curr_blk;
- unsigned char delete_blk_timer;
- unsigned short kactive_blk_num;
- unsigned short blk_sizeof_priv;
-
- /* last_kactive_blk_num:
- * trick to see if user-space has caught up
- * in order to avoid refreshing timer when every single pkt arrives.
- */
- unsigned short last_kactive_blk_num;
-
- char *pkblk_start;
- char *pkblk_end;
- int kblk_size;
- unsigned int knum_blocks;
- uint64_t knxt_seq_num;
- char *prev;
- char *nxt_offset;
- struct sk_buff *skb;
-
- atomic_t blk_fill_in_prog;
-
- /* Default is set to 8ms */
-#define DEFAULT_PRB_RETIRE_TOV (8)
-
- unsigned short retire_blk_tov;
- unsigned short version;
- unsigned long tov_in_jiffies;
-
- /* timer to retire an outstanding block */
- struct timer_list retire_blk_timer;
-};
-
#define PGV_FROM_VMALLOC 1
-struct pgv {
- char *buffer;
-};
-
-struct packet_ring_buffer {
- struct pgv *pg_vec;
- unsigned int head;
- unsigned int frames_per_block;
- unsigned int frame_size;
- unsigned int frame_max;
-
- unsigned int pg_vec_order;
- unsigned int pg_vec_pages;
- unsigned int pg_vec_len;
-
- struct tpacket_kbdq_core prb_bdqc;
- atomic_t pending;
-};
#define BLOCK_STATUS(x) ((x)->hdr.bh1.block_status)
#define BLOCK_NUM_PKTS(x) ((x)->hdr.bh1.num_pkts)
@@ -269,34 +207,6 @@ static void prb_fill_vlan_info(struct tpacket_kbdq_core *,
struct tpacket3_hdr *);
static void packet_flush_mclist(struct sock *sk);
-struct packet_fanout;
-struct packet_sock {
- /* struct sock has to be the first member of packet_sock */
- struct sock sk;
- struct packet_fanout *fanout;
- struct tpacket_stats stats;
- union tpacket_stats_u stats_u;
- struct packet_ring_buffer rx_ring;
- struct packet_ring_buffer tx_ring;
- int copy_thresh;
- spinlock_t bind_lock;
- struct mutex pg_vec_lock;
- unsigned int running:1, /* prot_hook is attached*/
- auxdata:1,
- origdev:1,
- has_vnet_hdr:1;
- int ifindex; /* bound device */
- __be16 num;
- struct packet_mclist *mclist;
- atomic_t mapped;
- enum tpacket_versions tp_version;
- unsigned int tp_hdrlen;
- unsigned int tp_reserve;
- unsigned int tp_loss:1;
- unsigned int tp_tstamp;
- struct packet_type prot_hook ____cacheline_aligned_in_smp;
-};
-
#define PACKET_FANOUT_MAX 256
struct packet_fanout {
@@ -334,11 +244,6 @@ struct packet_skb_cb {
(((x)->kactive_blk_num < ((x)->knum_blocks-1)) ? \
((x)->kactive_blk_num+1) : 0)
-static struct packet_sock *pkt_sk(struct sock *sk)
-{
- return (struct packet_sock *)sk;
-}
-
static void __fanout_unlink(struct sock *sk, struct packet_sock *po);
static void __fanout_link(struct sock *sk, struct packet_sock *po);
diff --git a/net/packet/internal.h b/net/packet/internal.h
new file mode 100644
index 0000000..2c5fca2
--- /dev/null
+++ b/net/packet/internal.h
@@ -0,0 +1,103 @@
+#ifndef __PACKET_INTERNAL_H__
+#define __PACKET_INTERNAL_H__
+
+struct packet_mclist {
+ struct packet_mclist *next;
+ int ifindex;
+ int count;
+ unsigned short type;
+ unsigned short alen;
+ unsigned char addr[MAX_ADDR_LEN];
+};
+
+/* kbdq - kernel block descriptor queue */
+struct tpacket_kbdq_core {
+ struct pgv *pkbdq;
+ unsigned int feature_req_word;
+ unsigned int hdrlen;
+ unsigned char reset_pending_on_curr_blk;
+ unsigned char delete_blk_timer;
+ unsigned short kactive_blk_num;
+ unsigned short blk_sizeof_priv;
+
+ /* last_kactive_blk_num:
+ * trick to see if user-space has caught up
+ * in order to avoid refreshing timer when every single pkt arrives.
+ */
+ unsigned short last_kactive_blk_num;
+
+ char *pkblk_start;
+ char *pkblk_end;
+ int kblk_size;
+ unsigned int knum_blocks;
+ uint64_t knxt_seq_num;
+ char *prev;
+ char *nxt_offset;
+ struct sk_buff *skb;
+
+ atomic_t blk_fill_in_prog;
+
+ /* Default is set to 8ms */
+#define DEFAULT_PRB_RETIRE_TOV (8)
+
+ unsigned short retire_blk_tov;
+ unsigned short version;
+ unsigned long tov_in_jiffies;
+
+ /* timer to retire an outstanding block */
+ struct timer_list retire_blk_timer;
+};
+
+struct pgv {
+ char *buffer;
+};
+
+struct packet_ring_buffer {
+ struct pgv *pg_vec;
+ unsigned int head;
+ unsigned int frames_per_block;
+ unsigned int frame_size;
+ unsigned int frame_max;
+
+ unsigned int pg_vec_order;
+ unsigned int pg_vec_pages;
+ unsigned int pg_vec_len;
+
+ struct tpacket_kbdq_core prb_bdqc;
+ atomic_t pending;
+};
+
+struct packet_fanout;
+struct packet_sock {
+ /* struct sock has to be the first member of packet_sock */
+ struct sock sk;
+ struct packet_fanout *fanout;
+ struct tpacket_stats stats;
+ union tpacket_stats_u stats_u;
+ struct packet_ring_buffer rx_ring;
+ struct packet_ring_buffer tx_ring;
+ int copy_thresh;
+ spinlock_t bind_lock;
+ struct mutex pg_vec_lock;
+ unsigned int running:1, /* prot_hook is attached*/
+ auxdata:1,
+ origdev:1,
+ has_vnet_hdr:1;
+ int ifindex; /* bound device */
+ __be16 num;
+ struct packet_mclist *mclist;
+ atomic_t mapped;
+ enum tpacket_versions tp_version;
+ unsigned int tp_hdrlen;
+ unsigned int tp_reserve;
+ unsigned int tp_loss:1;
+ unsigned int tp_tstamp;
+ struct packet_type prot_hook ____cacheline_aligned_in_smp;
+};
+
+static struct packet_sock *pkt_sk(struct sock *sk)
+{
+ return (struct packet_sock *)sk;
+}
+
+#endif
--
1.7.6.5
^ permalink raw reply related
* Re: [PATCH net-next] time: jiffies_delta_to_clock_t() helper to the rescue
From: Thomas Gleixner @ 2012-08-13 15:52 UTC (permalink / raw)
To: Eric Dumazet
Cc: David Miller, netdev, linux-kernel, Maciej Żenczykowski,
Paul Gortmaker, Andrew Morton, hank
In-Reply-To: <1344496433.28967.443.camel@edumazet-glaptop>
On Thu, 9 Aug 2012, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> Various /proc/net files sometimes report crazy timer values, expressed
> in clock_t units.
>
> This happens when an expired timer delta (expires - jiffies) is passed
> to jiffies_to_clock_t().
>
> This function has an overflow in :
>
> return div_u64((u64)x * TICK_NSEC, NSEC_PER_SEC / USER_HZ);
>
> commit cbbc719fccdb8cb (time: Change jiffies_to_clock_t() argument type
> to unsigned long) only got around the problem.
>
> As we cant output negative values in /proc/net/tcp without breaking
> various tools, I suggest adding a jiffies_delta_to_clock_t() wrapper
> that caps the negative delta to a 0 value.
That's correct for 64bit, but for 32bit you do the same for valid
timer values where expires < jiffies. So you might output 0 for valid
armed timers for quite a while around the point where jiffies are
close to the wrapping point.
Dunno if that's an issue, but it want's to be documented at least.
Thanks,
tglx
^ permalink raw reply
* [PATCH 2/4] packet: Diag core and basic socket info dumping
From: Pavel Emelyanov @ 2012-08-13 15:53 UTC (permalink / raw)
To: Linux Netdev List, David Miller
In-Reply-To: <5029219C.20305@parallels.com>
The diag module can be built independently from the af_packet.ko one,
just like it's done in unix sockets.
The core dumping message carries the info available at socket creation
time, i.e. family, type and protocol (in the same byte order as shown in
the proc file).
The socket inode number and cookie is reserved for future per-socket info
retrieving. The per-protocol filtering is also reserved for future by
requiring the sdiag_protocol to be zero.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
include/linux/Kbuild | 1 +
include/linux/packet_diag.h | 24 ++++++++++
net/packet/Kconfig | 8 +++
net/packet/Makefile | 2 +
net/packet/diag.c | 104 +++++++++++++++++++++++++++++++++++++++++++
5 files changed, 139 insertions(+), 0 deletions(-)
create mode 100644 include/linux/packet_diag.h
create mode 100644 net/packet/diag.c
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index d9a7544..d823d60 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -195,6 +195,7 @@ header-y += in_route.h
header-y += sock_diag.h
header-y += inet_diag.h
header-y += unix_diag.h
+header-y += packet_diag.h
header-y += inotify.h
header-y += input.h
header-y += ioctl.h
diff --git a/include/linux/packet_diag.h b/include/linux/packet_diag.h
new file mode 100644
index 0000000..fa19300
--- /dev/null
+++ b/include/linux/packet_diag.h
@@ -0,0 +1,24 @@
+#ifndef __PACKET_DIAG_H__
+#define __PACKET_DIAG_H__
+
+#include <linux/types.h>
+
+struct packet_diag_req {
+ __u8 sdiag_family;
+ __u8 sdiag_protocol;
+ __u16 pad;
+ __u32 pdiag_ino;
+ __u32 pdiag_show;
+ __u32 pdiag_cookie[2];
+};
+
+struct packet_diag_msg {
+ __u8 pdiag_family;
+ __u8 pdiag_type;
+ __u16 pdiag_num;
+
+ __u32 pdiag_ino;
+ __u32 pdiag_cookie[2];
+};
+
+#endif
diff --git a/net/packet/Kconfig b/net/packet/Kconfig
index 0060e3b..cc55b35 100644
--- a/net/packet/Kconfig
+++ b/net/packet/Kconfig
@@ -14,3 +14,11 @@ config PACKET
be called af_packet.
If unsure, say Y.
+
+config PACKET_DIAG
+ tristate "Packet: sockets monitoring interface"
+ depends on PACKET
+ default n
+ ---help---
+ Support for PF_PACKET sockets monitoring interface used by the ss tool.
+ If unsure, say Y.
diff --git a/net/packet/Makefile b/net/packet/Makefile
index 81183ea..9df6134 100644
--- a/net/packet/Makefile
+++ b/net/packet/Makefile
@@ -3,3 +3,5 @@
#
obj-$(CONFIG_PACKET) += af_packet.o
+obj-$(CONFIG_PACKET_DIAG) += af_packet_diag.o
+af_packet_diag-y += diag.o
diff --git a/net/packet/diag.c b/net/packet/diag.c
new file mode 100644
index 0000000..ff2f7f5
--- /dev/null
+++ b/net/packet/diag.c
@@ -0,0 +1,104 @@
+#include <linux/module.h>
+#include <linux/sock_diag.h>
+#include <linux/net.h>
+#include <linux/packet_diag.h>
+#include <net/net_namespace.h>
+#include <net/sock.h>
+
+#include "internal.h"
+
+static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct packet_diag_req *req,
+ u32 pid, u32 seq, u32 flags, int sk_ino)
+{
+ struct nlmsghdr *nlh;
+ struct packet_diag_msg *rp;
+ const struct packet_sock *po = pkt_sk(sk);
+
+ nlh = nlmsg_put(skb, pid, seq, SOCK_DIAG_BY_FAMILY, sizeof(*rp), flags);
+ if (!nlh)
+ return -EMSGSIZE;
+
+ rp = nlmsg_data(nlh);
+ rp->pdiag_family = AF_PACKET;
+ rp->pdiag_type = sk->sk_type;
+ rp->pdiag_num = ntohs(po->num);
+ rp->pdiag_ino = sk_ino;
+ sock_diag_save_cookie(sk, rp->pdiag_cookie);
+
+ return nlmsg_end(skb, nlh);
+}
+
+static int packet_diag_dump(struct sk_buff *skb, struct netlink_callback *cb)
+{
+ int num = 0, s_num = cb->args[0];
+ struct packet_diag_req *req;
+ struct net *net;
+ struct sock *sk;
+ struct hlist_node *node;
+
+ net = sock_net(skb->sk);
+ req = nlmsg_data(cb->nlh);
+
+ rcu_read_lock();
+ sk_for_each_rcu(sk, node, &net->packet.sklist) {
+ if (!net_eq(sock_net(sk), net))
+ continue;
+ if (num < s_num)
+ goto next;
+
+ if (sk_diag_fill(sk, skb, req, NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq, NLM_F_MULTI,
+ sock_i_ino(sk)) < 0)
+ goto done;
+next:
+ num++;
+ }
+done:
+ rcu_read_unlock();
+ cb->args[0] = num;
+
+ return skb->len;
+}
+
+static int packet_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h)
+{
+ int hdrlen = sizeof(struct packet_diag_req);
+ struct net *net = sock_net(skb->sk);
+ struct packet_diag_req *req;
+
+ if (nlmsg_len(h) < hdrlen)
+ return -EINVAL;
+
+ req = nlmsg_data(h);
+ /* Make it possible to support protocol filtering later */
+ if (req->sdiag_protocol)
+ return -EINVAL;
+
+ if (h->nlmsg_flags & NLM_F_DUMP) {
+ struct netlink_dump_control c = {
+ .dump = packet_diag_dump,
+ };
+ return netlink_dump_start(net->diag_nlsk, skb, h, &c);
+ } else
+ return -EOPNOTSUPP;
+}
+
+static const struct sock_diag_handler packet_diag_handler = {
+ .family = AF_PACKET,
+ .dump = packet_diag_handler_dump,
+};
+
+static int __init packet_diag_init(void)
+{
+ return sock_diag_register(&packet_diag_handler);
+}
+
+static void __exit packet_diag_exit(void)
+{
+ sock_diag_unregister(&packet_diag_handler);
+}
+
+module_init(packet_diag_init);
+module_exit(packet_diag_exit);
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_NETLINK, NETLINK_SOCK_DIAG, 17 /* AF_PACKET */);
--
1.7.6.5
^ permalink raw reply related
* [PATCH 2/4] packet: Report more packet sk info via diag module
From: Pavel Emelyanov @ 2012-08-13 15:55 UTC (permalink / raw)
To: Linux Netdev List, David Miller
In-Reply-To: <5029219C.20305@parallels.com>
This reports in one rtattr message all the other scalar values, that can be
set on a packet socket with setsockopt.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
include/linux/packet_diag.h | 23 +++++++++++++++++++++++
net/packet/diag.c | 33 +++++++++++++++++++++++++++++++++
2 files changed, 56 insertions(+), 0 deletions(-)
diff --git a/include/linux/packet_diag.h b/include/linux/packet_diag.h
index fa19300..3781ea4 100644
--- a/include/linux/packet_diag.h
+++ b/include/linux/packet_diag.h
@@ -12,6 +12,8 @@ struct packet_diag_req {
__u32 pdiag_cookie[2];
};
+#define PACKET_SHOW_INFO 0x00000001 /* Basic packet_sk information */
+
struct packet_diag_msg {
__u8 pdiag_family;
__u8 pdiag_type;
@@ -21,4 +23,25 @@ struct packet_diag_msg {
__u32 pdiag_cookie[2];
};
+enum {
+ PACKET_DIAG_INFO,
+
+ PACKET_DIAG_MAX,
+};
+
+struct packet_diag_info {
+ __u32 pdi_index;
+ __u32 pdi_version;
+ __u32 pdi_reserve;
+ __u32 pdi_copy_thresh;
+ __u32 pdi_tstamp;
+ __u32 pdi_flags;
+
+#define PDI_RUNNING 0x1
+#define PDI_AUXDATA 0x2
+#define PDI_ORIGDEV 0x4
+#define PDI_VNETHDR 0x8
+#define PDI_LOSS 0x10
+};
+
#endif
diff --git a/net/packet/diag.c b/net/packet/diag.c
index ff2f7f5..d5bd037 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -7,6 +7,31 @@
#include "internal.h"
+static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
+{
+ struct packet_diag_info pinfo;
+
+ pinfo.pdi_index = po->ifindex;
+ pinfo.pdi_version = po->tp_version;
+ pinfo.pdi_reserve = po->tp_reserve;
+ pinfo.pdi_copy_thresh = po->copy_thresh;
+ pinfo.pdi_tstamp = po->tp_tstamp;
+
+ pinfo.pdi_flags = 0;
+ if (po->running)
+ pinfo.pdi_flags |= PDI_RUNNING;
+ if (po->auxdata)
+ pinfo.pdi_flags |= PDI_AUXDATA;
+ if (po->origdev)
+ pinfo.pdi_flags |= PDI_ORIGDEV;
+ if (po->has_vnet_hdr)
+ pinfo.pdi_flags |= PDI_VNETHDR;
+ if (po->tp_loss)
+ pinfo.pdi_flags |= PDI_LOSS;
+
+ return nla_put(nlskb, PACKET_DIAG_INFO, sizeof(pinfo), &pinfo);
+}
+
static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct packet_diag_req *req,
u32 pid, u32 seq, u32 flags, int sk_ino)
{
@@ -25,7 +50,15 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct packet_diag
rp->pdiag_ino = sk_ino;
sock_diag_save_cookie(sk, rp->pdiag_cookie);
+ if ((req->pdiag_show & PACKET_SHOW_INFO) &&
+ pdiag_put_info(po, skb))
+ goto out_nlmsg_trim;
+
return nlmsg_end(skb, nlh);
+
+out_nlmsg_trim:
+ nlmsg_cancel(skb, nlh);
+ return -EMSGSIZE;
}
static int packet_diag_dump(struct sk_buff *skb, struct netlink_callback *cb)
--
1.7.6.5
^ permalink raw reply related
* [PATCH 4/4] packet: Report socket mclist info via diag module
From: Pavel Emelyanov @ 2012-08-13 15:57 UTC (permalink / raw)
To: Linux Netdev List, David Miller
In-Reply-To: <5029219C.20305@parallels.com>
The info is reported as an array of packet_diag_mclist structures. Each
includes not only the directly configured values (index, type, etc), but
also the "count".
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
include/linux/packet_diag.h | 10 ++++++++++
net/packet/diag.c | 39 +++++++++++++++++++++++++++++++++++++++
2 files changed, 49 insertions(+), 0 deletions(-)
diff --git a/include/linux/packet_diag.h b/include/linux/packet_diag.h
index 3781ea4..ea2e8923 100644
--- a/include/linux/packet_diag.h
+++ b/include/linux/packet_diag.h
@@ -13,6 +13,7 @@ struct packet_diag_req {
};
#define PACKET_SHOW_INFO 0x00000001 /* Basic packet_sk information */
+#define PACKET_SHOW_MCLIST 0x00000002 /* A set of packet_diag_mclist-s */
struct packet_diag_msg {
__u8 pdiag_family;
@@ -25,6 +26,7 @@ struct packet_diag_msg {
enum {
PACKET_DIAG_INFO,
+ PACKET_DIAG_MCLIST,
PACKET_DIAG_MAX,
};
@@ -44,4 +46,12 @@ struct packet_diag_info {
#define PDI_LOSS 0x10
};
+struct packet_diag_mclist {
+ __u32 pdmc_index;
+ __u32 pdmc_count;
+ __u16 pdmc_type;
+ __u16 pdmc_alen;
+ __u8 pdmc_addr[MAX_ADDR_LEN];
+};
+
#endif
diff --git a/net/packet/diag.c b/net/packet/diag.c
index d5bd037..3dda4ec 100644
--- a/net/packet/diag.c
+++ b/net/packet/diag.c
@@ -1,6 +1,7 @@
#include <linux/module.h>
#include <linux/sock_diag.h>
#include <linux/net.h>
+#include <linux/netdevice.h>
#include <linux/packet_diag.h>
#include <net/net_namespace.h>
#include <net/sock.h>
@@ -32,6 +33,40 @@ static int pdiag_put_info(const struct packet_sock *po, struct sk_buff *nlskb)
return nla_put(nlskb, PACKET_DIAG_INFO, sizeof(pinfo), &pinfo);
}
+static int pdiag_put_mclist(const struct packet_sock *po, struct sk_buff *nlskb)
+{
+ struct nlattr *mca;
+ struct packet_mclist *ml;
+
+ mca = nla_nest_start(nlskb, PACKET_DIAG_MCLIST);
+ if (!mca)
+ return -EMSGSIZE;
+
+ rtnl_lock();
+ for (ml = po->mclist; ml; ml = ml->next) {
+ struct packet_diag_mclist *dml;
+
+ dml = nla_reserve_nohdr(nlskb, sizeof(*dml));
+ if (!dml) {
+ rtnl_unlock();
+ nla_nest_cancel(nlskb, mca);
+ return -EMSGSIZE;
+ }
+
+ dml->pdmc_index = ml->ifindex;
+ dml->pdmc_type = ml->type;
+ dml->pdmc_alen = ml->alen;
+ dml->pdmc_count = ml->count;
+ BUILD_BUG_ON(sizeof(dml->pdmc_addr) != sizeof(ml->addr));
+ memcpy(dml->pdmc_addr, ml->addr, sizeof(ml->addr));
+ }
+
+ rtnl_unlock();
+ nla_nest_end(nlskb, mca);
+
+ return 0;
+}
+
static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct packet_diag_req *req,
u32 pid, u32 seq, u32 flags, int sk_ino)
{
@@ -54,6 +89,10 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct packet_diag
pdiag_put_info(po, skb))
goto out_nlmsg_trim;
+ if ((req->pdiag_show & PACKET_SHOW_MCLIST) &&
+ pdiag_put_mclist(po, skb))
+ goto out_nlmsg_trim;
+
return nlmsg_end(skb, nlh);
out_nlmsg_trim:
--
1.7.6.5
^ permalink raw reply related
* Re: [PATCH] ixgbevf - Remove unused parameter in ixgbevf_receive_skb
From: Waskiewicz Jr, Peter P @ 2012-08-13 15:58 UTC (permalink / raw)
To: Narendra_K@Dell.com; +Cc: netdev@vger.kernel.org, Kirsher, Jeffrey T
In-Reply-To: <20120813144255.GA4789@fedora-17-guest.blr.amer.dell.com>
On Mon, 2012-08-13 at 07:44 -0700, Narendra_K@Dell.com wrote:
> From: Narendra K <narendra_k@dell.com>
>
> Remove 'rx_ring' parameter as it is not used in ixgbevf_receive_skb
>
> Signed-off-by: Narendra K <narendra_k@dell.com>
> ---
> The patch applies to 'net' tree.
Thanks Narenda, we'll get it applied and tested in our internal trees.
Cheers,
-PJ
^ permalink raw reply
* Re: [PATCH V2 09/12] net/eipoib: Add main driver functionality
From: Eric W. Biederman @ 2012-08-13 16:08 UTC (permalink / raw)
To: Or Gerlitz
Cc: Michael S. Tsirkin, davem, roland, netdev, ali, sean.hefty,
Erez Shitrit, Doug Ledford
In-Reply-To: <5028BBCE.4020908@mellanox.com>
Or Gerlitz <ogerlitz@mellanox.com> wrote:
>On 12/08/2012 18:40, Eric W. Biederman wrote:
>> Let me give you a non-hack recomendation.
>>
>> - Give up on being wire compatible with IPoIB.
>>
>> - Define and implement ethernet over inifiniband aka EoIB.
>>
>> With EoIB:
>> - The SM would map ethernet address to inifiniband hardware
>addresses.
>> - You discover which multicast addresses are of interest from the
>> IP layer above so no snooping is necessary.
>> - You could run queue pairs directly to hosts.
>>
>> Shrug. It is trivial and it will work. It will probably run into
>the
>> same problems that have historically been a problem for using IPoIB
>> (lack of stateless offloads) but shrug that is mostly a NIC firmware
>> problem. The switches will have no trouble and interoperability will
>> be assured.
>>
>> If you want to map ethernet over infiniband please map ethernet over
>> infiniband. Don't poorly NAT ethernet into infiniband.
>>
>>
>
>
>EoIB is a valid suggestion and we will look into it as well, BUT:
>
>Providing EoIB is a separate discussion, obviously defining and
>standardizing a new protocol takes what is takes (a lot of time,
>longish
>term effort), and will also take time to develop/debug/mature e.g as
If you follow Michael Tirskins suggestion and use the same wire encoding as IPoIB and infer the mac address from the lids and queue pair numbers as you are already doing with for eIPoIB, except for defining exactly how to get the subnet manager to store the mac address to lid/qpn mapping you are done.
If you don't involve a comitte and simply define a defacto standard it will take less effort than this conversation and less effort than implementing your eIPoIB driver.
>you
>mentioned, some of the features/offloads might require new NIC HW, etc
>-- compared to IPoIB which is here for many years
So deploy routing and proxy arp and you are done.
>In practice there is already a huge install base for IPoIB software and
>
>hardware products, in different operating environments/OS. We can't
>just
>through away everything and tell people to replace it all with a new
>protocol, e.g. bridging devices, storage systems/appliances, VMware,
>Windows, .. systems in production environments --- so
>the interoperability concern you've mentioned gonna hit very hard.
There is no need to throw anything away. Just put them on different IP subnets.
Shrug.
>The eIPoIB driver comes to provide a way to work with IPoIB in
>virtualized environments, where still, the suggestions/concerns raised
>in this thread should be addressed.
eIPoIB does not work.
I can't get an IP address with out a specially configured dhcp server, and special dhcp clients.
eIPoIB does not work with IPv6.
As David Miller already said this code has no chance of being merged.
Shrug. I have been polite and pointed out implementation choices that actually work. Solutions that are less effort and less code, and provide more interoperability.
If after patient explanation you can not appreciate why people consider eIPoIB to be totally unacceptable that is your problem.
Good luck in your future endeavours,
Eric
^ permalink raw reply
* Re: [PATCH] configure: Add search path for 64bit library.
From: Ben Hutchings @ 2012-08-13 16:22 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: Li Wei, netdev
In-Reply-To: <20120813082624.153167a8@nehalam.linuxnetplumber.net>
On Mon, 2012-08-13 at 08:26 -0700, Stephen Hemminger wrote:
> On Tue, 7 Aug 2012 19:15:58 +0100
> Ben Hutchings <bhutchings@solarflare.com> wrote:
>
> > The subject line doesn't say what this is for, but it looks like
> > iproute2...
> >
> > On Tue, 2012-08-07 at 12:22 +0800, Li Wei wrote:
> > > Signed-off-by: Li Wei <lw@cn.fujitsu.com>
> > > ---
> > > configure | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/configure b/configure
> > > index 0f4444f..997759c 100755
> > > --- a/configure
> > > +++ b/configure
> > > @@ -149,7 +149,7 @@ check_ipt()
> > > check_ipt_lib_dir()
> > > {
> > > IPT_LIB_DIR=""
> > > - for dir in /lib /usr/lib /usr/local/lib
> > > + for dir in /lib /usr/lib /usr/local/lib /lib64 /usr/lib64 /usr/local/lib64
> > > do
> > > for file in $dir/{xtables,iptables}/lib*t_*so ; do
> > > if [ -f $file ]; then
> >
> > On a bi-arch system, surely the lib64 directories should be preferred to
> > the lib directories? And this still leaves multi-arch to be handled.
> >
> > I think this should be done with pkg-config:
> >
> > pkg-config --variable=xtlibdir xtables
> >
> > possibly with that directory list as a fallback if it's useful to
> > support iptables library versions that didn't include xtables.pc.
> >
> > Ben.
> >
>
> Does every distro have pkg-config or does more logic need to be done here?
Every distro has pkg-config; the question is whether you want to support
library versions that don't include a pkg-config file (xtables.pc), if
they exist.
Ben.
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [RFC net-next 0/4] gianfar: Use separate NAPI for Tx confirmation processing
From: Claudiu Manoil @ 2012-08-13 16:23 UTC (permalink / raw)
To: Tomas Hruby, Eric Dumazet, Paul Gortmaker; +Cc: netdev, David S. Miller
In-Reply-To: <5023D21E.1000008@freescale.com>
On 08/09/2012 06:07 PM, Claudiu Manoil wrote:
> On 8/9/2012 2:06 AM, Tomas Hruby wrote:
>> On Wed, Aug 8, 2012 at 9:44 AM, Eric Dumazet <eric.dumazet@gmail.com>
>> wrote:
>>> On Wed, 2012-08-08 at 12:24 -0400, Paul Gortmaker wrote:
>>>> [[RFC net-next 0/4] gianfar: Use separate NAPI for Tx confirmation
>>>> processing] On 08/08/2012 (Wed 15:26) Claudiu Manoil wrote:
>>>>
>>>>> Hi all,
>>>>> This set of patches basically splits the existing napi poll
>>>>> routine into
>>>>> two separate napi functions, one for Rx processing (triggered by
>>>>> frame
>>>>> receive interrupts only) and one for the Tx confirmation path
>>>>> processing
>>>>> (triggerred by Tx confirmation interrupts only). The polling
>>>>> algorithm
>>>>> behind remains much the same.
>>>>>
>>>>> Important throughput improvements have been noted on low power
>>>>> boards with
>>>>> this set of changes.
>>>>> For instance, for the following netperf test:
>>>>> netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
>>>>> yields a throughput gain from oscilating ~500-~700 Mbps to steady
>>>>> ~940 Mbps,
>>>>> (if the Rx/Tx paths are processed on different cores), w/ no
>>>>> increase in CPU%,
>>>>> on a p1020rdb - 2 core machine featuring etsec2.0 (Multi-Queue
>>>>> Multi-Group
>>>>> driver mode).
>>>>
>>>> It would be interesting to know more about what was causing that large
>>>> an oscillation -- presumably you will have it reappear once one core
>>>> becomes 100% utilized. Also, any thoughts on how the change will
>>>> change
>>>> performance on an older low power single core gianfar system (e.g.
>>>> 83xx)?
>>>
>>> I also was wondering if this low performance could be caused by BQL
>>>
>>> Since TCP stack is driven by incoming ACKS, a NAPI run could have to
>>> handle 10 TCP acks in a row, and resulting xmits could hit BQL and
>>> transit on qdisc (Because NAPI handler wont handle TX completions in
>>> the
>>> middle of RX handler)
>>
>> Does disabling BQL help? Is the BQL limit stable? To what value is it
>> set? I would be very much interested in more data if the issue is BQL
>> related.
>>
>> .
>>
>
> I agree that more tests should be run to investigate why gianfar under-
> performs on the low power p1020rdb platform, and BQL seems to be
> a good starting point (thanks for the hint). What I can say now is that
> the issue is not apparent on p2020rdb, for instance, which is a more
> powerful platform: the CPUs - 1200 MHz instead of 800 MHz; twice the
> size of L2 cache (512 KB), greater bus (CCB) frequency ... On this
> board (p2020rdb) the netperf test reaches 940Mbps both w/ and w/o these
> patches.
>
> For a single core system I'm not expecting any performance degradation,
> simply because I don't see why the proposed napi poll implementation
> would be slower than the existing one. I'll do some measurements on a
> p1010rdb too (single core, CPU:800 MHz) and get back to you with the
> results.
>
Hi all,
Please find below the netperf measurements performed on a p1010rdb machine
(single core, low power). Three kernel images were used:
1) Linux version 3.5.0-20970-gaae06bf -- net-next commit aae06bf
2) Linux version 3.5.0-20974-g2920464 -- commit aae06bf + Tx NAPI patches
3) Linux version 3.5.0-20970-gaae06bf-dirty -- commit aae06bf +
CONFIG_BQL set to 'n'
The results show that, on *Image 1)*, by adjusting
tcp_limit_output_bytes no substantial
improvements are seen, as the throughput stays in the 580-60x Mbps range .
By changing the coalescing settings from default* (rx coalescing off,
tx-usecs: 10, tx-frames: 16) to:
"ethtool -C eth1 rx-frames 22 tx-frames 22 rx-usecs 32 tx-usecs 32"
we get a throughput of ~710 Mbps.
For *Image 2)*, using the default tcp_limit_output_bytes value (131072)
- I've noticed
that "tweaking" tcp_limit_output_bytes does not improve the throughput
-, we get the
following performance numbers:
* default coalescing settings: ~650 Mbps
* rx-frames tx-frames 22 rx-usecs 32 tx-usecs 32: ~860-880 Mbps
For *Image 3)*, by disabling BQL (CONFIG_BQL = n), there's *no* relevant
performance
improvement compared to Image 1).
(note:
For all the measurements, rx and tx BD ring sizes have been set to 64,
for best performance.)
So, I really tend to believe that the performance degradation comes
primarily from the driver,
and the napi poll processing turns out to be an important source for
that. The proposed patches
show substantial improvement, especially for SMP systems where Tx and Rx
processing may be
done in parallel.
What do you think?
Is it ok to proceed by re-spinning the patches? Do you recommend
additional measurements?
Regards,
Claudiu
//=Image 1)================
root@p1010rdb:~# cat /proc/version
Linux version 3.5.0-20970-gaae06bf [...]
root@p1010rdb:~# zcat /proc/config.gz | grep BQL
CONFIG_BQL=y
root@p1010rdb:~# cat /proc/sys/net/ipv4/tcp_limit_output_bytes
131072
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 580.76 99.95 11.76 14.099 1.659
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 598.21 99.95 10.91 13.687 1.493
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 583.04 99.95 11.25 14.043 1.581
root@p1010rdb:~# cat /proc/sys/net/ipv4/tcp_limit_output_bytes
65536
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 604.29 99.95 11.15 13.550 1.512
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 603.52 99.50 12.57 13.506 1.706
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 596.18 99.95 12.81 13.734 1.760
root@p1010rdb:~# cat /proc/sys/net/ipv4/tcp_limit_output_bytes
32768
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 582.32 99.95 12.96 14.061 1.824
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 583.79 99.95 11.19 14.026 1.571
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 584.16 99.95 11.36 14.016 1.592
root@p1010rdb:~# ethtool -C eth1 rx-frames 22 tx-frames 22 rx-usecs 32
tx-usecs 32
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 708.77 99.85 13.32 11.541 1.540
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 710.50 99.95 12.46 11.524 1.437
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 709.95 99.95 14.15 11.533 1.633
//=Image 2)================
root@p1010rdb:~# cat /proc/version
Linux version 3.5.0-20974-g2920464 [...]
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 652.60 99.95 13.05 12.547 1.638
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 657.47 99.95 11.81 12.454 1.471
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 655.77 99.95 11.80 12.486 1.474
root@p1010rdb:~# ethtool -C eth1 rx-frames 22 rx-usecs 32 tx-frames 22
tx-usecs 32
root@p1010rdb:~# cat /proc/sys/net/ipv4/tcp_limit_output_bytes
131072
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.01 882.42 99.20 18.06 9.209 1.676
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 867.02 99.75 16.21 9.425 1.531
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.01 874.29 99.85 15.25 9.356 1.429
//=Image 3)================
Linux version 3.5.0-20970-gaae06bf-dirty [...] //CONFIG_BQL = n
root@p1010rdb:~# cat /proc/version
Linux version 3.5.0-20970-gaae06bf-dirty
(b08782@zro04-ws574.ea.freescale.net) (gcc version 4.6.2 (GCC) ) #3 Mon
Aug 13 13:58:25 EEST 2012
root@p1010rdb:~# zcat /proc/config.gz | grep BQL
# CONFIG_BQL is not set
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 595.08 99.95 12.51 13.759 1.722
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 593.95 99.95 10.96 13.785 1.511
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 595.30 99.90 11.11 13.747 1.528
root@p1010rdb:~# ethtool -C eth1 rx-frames 22 rx-usecs 32 tx-frames 22
tx-usecs 32
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 710.46 99.95 12.46 11.525 1.437
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 714.27 99.95 14.05 11.463 1.611
root@p1010rdb:~# netperf -l 20 -cC -H 192.168.10.1 -t TCP_STREAM -- -m 1500
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.10.1
(192.168.10.1) port 0 AF_INET
Recv Send Send Utilization Service Demand
Socket Socket Message Elapsed Send Recv Send Recv
Size Size Size Time Throughput local remote local remote
bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB
87380 16384 1500 20.00 717.69 99.95 12.56 11.409 1.433
.
^ permalink raw reply
* Re: [PATCH] configure: Add search path for 64bit library.
From: Stephen Hemminger @ 2012-08-13 16:33 UTC (permalink / raw)
To: Ben Hutchings; +Cc: Li Wei, netdev
In-Reply-To: <1344874960.2733.2.camel@bwh-desktop.uk.solarflarecom.com>
On Mon, 13 Aug 2012 17:22:40 +0100
Ben Hutchings <bhutchings@solarflare.com> wrote:
> On Mon, 2012-08-13 at 08:26 -0700, Stephen Hemminger wrote:
> > On Tue, 7 Aug 2012 19:15:58 +0100
> > Ben Hutchings <bhutchings@solarflare.com> wrote:
> >
> > > The subject line doesn't say what this is for, but it looks like
> > > iproute2...
> > >
> > > On Tue, 2012-08-07 at 12:22 +0800, Li Wei wrote:
> > > > Signed-off-by: Li Wei <lw@cn.fujitsu.com>
> > > > ---
> > > > configure | 2 +-
> > > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > > >
> > > > diff --git a/configure b/configure
> > > > index 0f4444f..997759c 100755
> > > > --- a/configure
> > > > +++ b/configure
> > > > @@ -149,7 +149,7 @@ check_ipt()
> > > > check_ipt_lib_dir()
> > > > {
> > > > IPT_LIB_DIR=""
> > > > - for dir in /lib /usr/lib /usr/local/lib
> > > > + for dir in /lib /usr/lib /usr/local/lib /lib64 /usr/lib64 /usr/local/lib64
> > > > do
> > > > for file in $dir/{xtables,iptables}/lib*t_*so ; do
> > > > if [ -f $file ]; then
> > >
> > > On a bi-arch system, surely the lib64 directories should be preferred to
> > > the lib directories? And this still leaves multi-arch to be handled.
> > >
> > > I think this should be done with pkg-config:
> > >
> > > pkg-config --variable=xtlibdir xtables
> > >
> > > possibly with that directory list as a fallback if it's useful to
> > > support iptables library versions that didn't include xtables.pc.
> > >
> > > Ben.
> > >
> >
> > Does every distro have pkg-config or does more logic need to be done here?
>
> Every distro has pkg-config; the question is whether you want to support
> library versions that don't include a pkg-config file (xtables.pc), if
> they exist.
Let's do pkg-config first, and as a fallback keep the old code and only
look in the same old places.
^ permalink raw reply
* Re: [RFC PATCH 0/2] net: connect to UNIX sockets from specified root
From: J. Bruce Fields @ 2012-08-13 16:47 UTC (permalink / raw)
To: Stanislav Kinsbursky
Cc: Pavel Emelyanov, H. Peter Anvin, Alan Cox,
Trond.Myklebust@netapp.com, davem@davemloft.net,
linux-nfs@vger.kernel.org, eric.dumazet@gmail.com,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
viro@zeniv.linux.org.uk, tim.c.chen@linux.intel.com,
devel@openvz.org
In-Reply-To: <50263ECC.4060501@parallels.com>
On Sat, Aug 11, 2012 at 03:15:24PM +0400, Stanislav Kinsbursky wrote:
> 11.08.2012 10:23, Pavel Emelyanov пишет:
> >On 08/11/2012 03:09 AM, H. Peter Anvin wrote:
> >>On 08/10/2012 12:28 PM, Alan Cox wrote:
> >>>Explicitly for Linux yes - this is not generally true of the AF_UNIX
> >>>socket domain and even the permissions aspect isn't guaranteed to be
> >>>supported on some BSD environments !
> >>Yes, but let's worry about what the Linux behavior should be.
> >>
> >>>The name is however just a proxy for the socket itself. You don't even
> >>>get a device node in the usual sense or the same inode in the file system
> >>>space.
> >>
> >>No, but it is looked up the same way any other inode is (the difference
> >>between FIFOs and sockets is that sockets have separate connections,
> >>which is also why open() on sockets would be nice.)
> >>
> >>However, there is a fundamental difference between AF_UNIX sockets and
> >>open(), and that is how the pathname is delivered. It thus would make
> >>more sense to provide the openat()-like information in struct
> >>sockaddr_un, but that may be very hard to do in a sensible way. In that
> >>sense it perhaps would be cleaner to be able to do an open[at]() on the
> >>socket node with O_PATH (perhaps there should be an O_SOCKET option,
> >>even?) and pass the resulting file descriptor to bind() or connect().
> >I vote for this (openat + O_WHATEVER on a unix socket) as well. It will
> >help us in checkpoint-restore, making handling of overmounted/unlinked
> >sockets much cleaner.
>
> I have to notice, that it's not enough and doesn't solve the issue.
> There should be some way how to connect/bind already existent unix
> socket (from kernel, at least), because socket can be created in
> user space.
> And this way (sock operation or whatever) have to provide an ability
> to lookup UNIX socket starting from specified root to support
> containers.
I don't understand--the rpcbind sockets are created by the kernel. What
am I missing?
--b.
^ permalink raw reply
* Re: [flame^Wreview] net: netprio_cgroup: rework update socket logic
From: John Fastabend @ 2012-08-13 16:58 UTC (permalink / raw)
To: Al Viro; +Cc: netdev, David Miller, Neil Horman, linux-kernel
In-Reply-To: <20120813121827.GB23464@ZenIV.linux.org.uk>
[...]
> HOWEVER, it still doesn't address more fundamental problem - somebody
> creating a socket and passing it to you in SCM_RIGHTS datagram will
> leave you with a socket you can do IO on, still tagged according to who
> had created it.
>
> AFAICS, the whole point of that exercise was to allow third-party changing
> the priorities of traffic on sockets already created by a process we now
> move to a different cgroup. Consider e.g. this:
Correct that is the point of the exercise.
To fix this specific case we could add a call to sock_update_netprioidx
in scm_recv to set the sk_cgrp_prioidx value.
^ permalink raw reply
* Re: [flame^Wreview] net: netprio_cgroup: rework update socket logic
From: Al Viro @ 2012-08-13 17:01 UTC (permalink / raw)
To: John Fastabend; +Cc: netdev, David Miller, Neil Horman, linux-kernel
In-Reply-To: <50293224.90803@intel.com>
On Mon, Aug 13, 2012 at 09:58:12AM -0700, John Fastabend wrote:
> [...]
>
> >HOWEVER, it still doesn't address more fundamental problem - somebody
> >creating a socket and passing it to you in SCM_RIGHTS datagram will
> >leave you with a socket you can do IO on, still tagged according to who
> >had created it.
> >
> >AFAICS, the whole point of that exercise was to allow third-party changing
> >the priorities of traffic on sockets already created by a process we now
> >move to a different cgroup. Consider e.g. this:
>
> Correct that is the point of the exercise.
>
> To fix this specific case we could add a call to sock_update_netprioidx
> in scm_recv to set the sk_cgrp_prioidx value.
On every received descriptor, that is? Eeek...
^ permalink raw reply
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