* [ofa-general] [PATCH V3 2/7] IB/ipoib: Verify address handle validity on send
From: Moni Shoua @ 2007-07-30 12:49 UTC (permalink / raw)
To: rdreier, davem, fubar; +Cc: netdev, general
In-Reply-To: <46ADDB89.5030601@voltaire.com>
When the bonding device senses a carrier loss of its active slave it replaces
that slave with a new one. In between the times when the carrier of an IPoIB
device goes down and ipoib_neigh is destroyed, it is possible that the
bonding driver will send a packet on a new slave that uses an old ipoib_neigh.
This patch detects and prevents this from happenning.
Signed-off-by: Moni Shoua <monis@voltaire.com>
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
---
drivers/infiniband/ulp/ipoib/ipoib_main.c | 5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
Index: net-2.6/drivers/infiniband/ulp/ipoib/ipoib_main.c
===================================================================
--- net-2.6.orig/drivers/infiniband/ulp/ipoib/ipoib_main.c 2007-07-25 14:57:48.000000000 +0300
+++ net-2.6/drivers/infiniband/ulp/ipoib/ipoib_main.c 2007-07-25 15:02:55.525131034 +0300
@@ -685,9 +685,10 @@ static int ipoib_start_xmit(struct sk_bu
goto out;
}
} else if (neigh->ah) {
- if (unlikely(memcmp(&neigh->dgid.raw,
+ if (unlikely((memcmp(&neigh->dgid.raw,
skb->dst->neighbour->ha + 4,
- sizeof(union ib_gid)))) {
+ sizeof(union ib_gid))) ||
+ (neigh->dev != dev))) {
spin_lock(&priv->lock);
/*
* It's safe to call ipoib_put_ah() inside
^ permalink raw reply
* [ofa-general] [PATCH V3 3/7] net/bonding: Enable bonding to enslave non ARPHRD_ETHER
From: Moni Shoua @ 2007-07-30 12:51 UTC (permalink / raw)
To: rdreier, davem, fubar; +Cc: netdev, general
In-Reply-To: <46ADDB89.5030601@voltaire.com>
This patch changes some of the bond netdevice attributes and functions
to be that of the active slave for the case of the enslaved device not being
of ARPHRD_ETHER type. Basically it overrides those setting done by ether_setup(),
which are netdevice **type** dependent and hence might be not appropriate for
devices of other types. It also enforces mutual exclusion on bonding slaves
from dissimilar ether types, as was concluded over the v1 discussion.
IPoIB (see Documentation/infiniband/ipoib.txt) MAC address is made of a 3 bytes
IB QP (Queue Pair) number and 16 bytes IB port GID (Global ID) of the port this
IPoIB device is bounded to. The QP is a resource created by the IB HW and the
GID is an identifier burned into the HCA (i have omitted here some details which
are not important for the bonding RFC).
Signed-off-by: Moni Shoua <monis@voltaire.com>
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
---
drivers/net/bonding/bond_main.c | 38 ++++++++++++++++++++++++++++++++++++++
1 files changed, 38 insertions(+)
Index: net-2.6/drivers/net/bonding/bond_main.c
===================================================================
--- net-2.6.orig/drivers/net/bonding/bond_main.c 2007-07-25 15:02:10.000000000 +0300
+++ net-2.6/drivers/net/bonding/bond_main.c 2007-07-29 16:24:30.913343981 +0300
@@ -1277,6 +1277,26 @@ static int bond_compute_features(struct
return 0;
}
+
+static void bond_setup_by_slave(struct net_device *bond_dev,
+ struct net_device *slave_dev)
+{
+ bond_dev->hard_header = slave_dev->hard_header;
+ bond_dev->rebuild_header = slave_dev->rebuild_header;
+ bond_dev->hard_header_cache = slave_dev->hard_header_cache;
+ bond_dev->header_cache_update = slave_dev->header_cache_update;
+ bond_dev->hard_header_parse = slave_dev->hard_header_parse;
+
+ bond_dev->neigh_setup = slave_dev->neigh_setup;
+
+ bond_dev->type = slave_dev->type;
+ bond_dev->hard_header_len = slave_dev->hard_header_len;
+ bond_dev->addr_len = slave_dev->addr_len;
+
+ memcpy(bond_dev->broadcast, slave_dev->broadcast,
+ slave_dev->addr_len);
+}
+
/* enslave device <slave> to bond device <master> */
int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
{
@@ -1351,6 +1371,24 @@ int bond_enslave(struct net_device *bond
goto err_undo_flags;
}
+ /* set bonding device ether type by slave - bonding netdevices are
+ * created with ether_setup, so when the slave type is not ARPHRD_ETHER
+ * there is a need to override some of the type dependent attribs/funcs.
+ *
+ * bond ether type mutual exclusion - don't allow slaves of dissimilar
+ * ether type (eg ARPHRD_ETHER and ARPHRD_INFINIBAND) share the same bond
+ */
+ if (bond->slave_cnt == 0) {
+ if (slave_dev->type != ARPHRD_ETHER)
+ bond_setup_by_slave(bond_dev, slave_dev);
+ } else if (bond_dev->type != slave_dev->type) {
+ printk(KERN_ERR DRV_NAME ": %s ether type (%d) is different from "
+ "other slaves (%d), can not enslave it.\n", slave_dev->name,
+ slave_dev->type, bond_dev->type);
+ res = -EINVAL;
+ goto err_undo_flags;
+ }
+
if (slave_dev->set_mac_address == NULL) {
printk(KERN_ERR DRV_NAME
": %s: Error: The slave device you specified does "
^ permalink raw reply
* [ofa-general] [PATCH V3 4/7] net/bonding: Enable bonding to enslave netdevices not supporting set_mac_address()
From: Moni Shoua @ 2007-07-30 12:52 UTC (permalink / raw)
To: rdreier, davem, fubar; +Cc: netdev, general
In-Reply-To: <46ADDB89.5030601@voltaire.com>
This patch allows for enslaving netdevices which do not support
the set_mac_address() function. In that case the bond mac address is the one
of the active slave, where remote peers are notified on the mac address
(neighbour) change by Gratuitous ARP sent by bonding when fail-over occurs
(this is already done by the bonding code).
Signed-off-by: Moni Shoua <monis@voltaire.com>
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
---
drivers/net/bonding/bond_main.c | 88 +++++++++++++++++++++++++++-------------
drivers/net/bonding/bonding.h | 1
2 files changed, 61 insertions(+), 28 deletions(-)
Index: net-2.6/drivers/net/bonding/bond_main.c
===================================================================
--- net-2.6.orig/drivers/net/bonding/bond_main.c 2007-07-29 16:24:30.913343981 +0300
+++ net-2.6/drivers/net/bonding/bond_main.c 2007-07-29 16:36:53.234602471 +0300
@@ -1127,6 +1127,14 @@ void bond_change_active_slave(struct bon
if (new_active) {
bond_set_slave_active_flags(new_active);
}
+
+ /* when bonding does not set the slave MAC address, the bond MAC
+ * address is the one of the active slave.
+ */
+ if (new_active && !bond->do_set_mac_addr)
+ memcpy(bond->dev->dev_addr, new_active->dev->dev_addr,
+ new_active->dev->addr_len);
+
bond_send_gratuitous_arp(bond);
}
}
@@ -1390,13 +1398,22 @@ int bond_enslave(struct net_device *bond
}
if (slave_dev->set_mac_address == NULL) {
- printk(KERN_ERR DRV_NAME
- ": %s: Error: The slave device you specified does "
- "not support setting the MAC address. "
- "Your kernel likely does not support slave "
- "devices.\n", bond_dev->name);
- res = -EOPNOTSUPP;
- goto err_undo_flags;
+ if (bond->slave_cnt == 0) {
+ printk(KERN_WARNING DRV_NAME
+ ": %s: Warning: The first slave device you "
+ "specified does not support setting the MAC "
+ "address. This bond MAC address would be that "
+ "of the active slave.\n", bond_dev->name);
+ bond->do_set_mac_addr = 0;
+ } else if (bond->do_set_mac_addr) {
+ printk(KERN_ERR DRV_NAME
+ ": %s: Error: The slave device you specified "
+ "does not support setting the MAC addres,."
+ "but this bond uses this practice. \n"
+ , bond_dev->name);
+ res = -EOPNOTSUPP;
+ goto err_undo_flags;
+ }
}
new_slave = kzalloc(sizeof(struct slave), GFP_KERNEL);
@@ -1417,16 +1434,18 @@ int bond_enslave(struct net_device *bond
*/
memcpy(new_slave->perm_hwaddr, slave_dev->dev_addr, ETH_ALEN);
- /*
- * Set slave to master's mac address. The application already
- * set the master's mac address to that of the first slave
- */
- memcpy(addr.sa_data, bond_dev->dev_addr, bond_dev->addr_len);
- addr.sa_family = slave_dev->type;
- res = dev_set_mac_address(slave_dev, &addr);
- if (res) {
- dprintk("Error %d calling set_mac_address\n", res);
- goto err_free;
+ if (bond->do_set_mac_addr) {
+ /*
+ * Set slave to master's mac address. The application already
+ * set the master's mac address to that of the first slave
+ */
+ memcpy(addr.sa_data, bond_dev->dev_addr, bond_dev->addr_len);
+ addr.sa_family = slave_dev->type;
+ res = dev_set_mac_address(slave_dev, &addr);
+ if (res) {
+ dprintk("Error %d calling set_mac_address\n", res);
+ goto err_free;
+ }
}
res = netdev_set_master(slave_dev, bond_dev);
@@ -1651,9 +1670,11 @@ err_close:
dev_close(slave_dev);
err_restore_mac:
- memcpy(addr.sa_data, new_slave->perm_hwaddr, ETH_ALEN);
- addr.sa_family = slave_dev->type;
- dev_set_mac_address(slave_dev, &addr);
+ if (bond->do_set_mac_addr) {
+ memcpy(addr.sa_data, new_slave->perm_hwaddr, ETH_ALEN);
+ addr.sa_family = slave_dev->type;
+ dev_set_mac_address(slave_dev, &addr);
+ }
err_free:
kfree(new_slave);
@@ -1831,10 +1852,12 @@ int bond_release(struct net_device *bond
/* close slave before restoring its mac address */
dev_close(slave_dev);
- /* restore original ("permanent") mac address */
- memcpy(addr.sa_data, slave->perm_hwaddr, ETH_ALEN);
- addr.sa_family = slave_dev->type;
- dev_set_mac_address(slave_dev, &addr);
+ if (bond->do_set_mac_addr) {
+ /* restore original ("permanent") mac address */
+ memcpy(addr.sa_data, slave->perm_hwaddr, ETH_ALEN);
+ addr.sa_family = slave_dev->type;
+ dev_set_mac_address(slave_dev, &addr);
+ }
slave_dev->priv_flags &= ~(IFF_MASTER_8023AD | IFF_MASTER_ALB |
IFF_SLAVE_INACTIVE | IFF_BONDING |
@@ -1921,10 +1944,12 @@ static int bond_release_all(struct net_d
/* close slave before restoring its mac address */
dev_close(slave_dev);
- /* restore original ("permanent") mac address*/
- memcpy(addr.sa_data, slave->perm_hwaddr, ETH_ALEN);
- addr.sa_family = slave_dev->type;
- dev_set_mac_address(slave_dev, &addr);
+ if (bond->do_set_mac_addr) {
+ /* restore original ("permanent") mac address*/
+ memcpy(addr.sa_data, slave->perm_hwaddr, ETH_ALEN);
+ addr.sa_family = slave_dev->type;
+ dev_set_mac_address(slave_dev, &addr);
+ }
slave_dev->priv_flags &= ~(IFF_MASTER_8023AD | IFF_MASTER_ALB |
IFF_SLAVE_INACTIVE);
@@ -3961,6 +3986,10 @@ static int bond_set_mac_address(struct n
dprintk("bond=%p, name=%s\n", bond, (bond_dev ? bond_dev->name : "None"));
+ if (!bond->do_set_mac_addr) {
+ return -EOPNOTSUPP;
+ }
+
if (!is_valid_ether_addr(sa->sa_data)) {
return -EADDRNOTAVAIL;
}
@@ -4351,6 +4380,9 @@ static int bond_init(struct net_device *
bond_create_proc_entry(bond);
#endif
+ /* set do_set_mac_addr to true on startup */
+ bond->do_set_mac_addr = 1;
+
list_add_tail(&bond->bond_list, &bond_dev_list);
return 0;
Index: net-2.6/drivers/net/bonding/bonding.h
===================================================================
--- net-2.6.orig/drivers/net/bonding/bonding.h 2007-07-29 16:25:22.000000000 +0300
+++ net-2.6/drivers/net/bonding/bonding.h 2007-07-29 16:37:13.163056181 +0300
@@ -201,6 +201,7 @@ struct bonding {
struct list_head vlan_list;
struct vlan_group *vlgrp;
struct packet_type arp_mon_pt;
+ s8 do_set_mac_addr;
};
/**
^ permalink raw reply
* [ofa-general] [PATCH V3 5/7] net/bonding: Enable IP multicast for bonding IPoIB devices
From: Moni Shoua @ 2007-07-30 12:54 UTC (permalink / raw)
To: rdreier, davem, fubar; +Cc: netdev, general
In-Reply-To: <46ADDB89.5030601@voltaire.com>
Allow to enslave devices when the bonding device is not up. Over the discussion
held at the previous post this seemed to be the most clean way to go, where it
is not expected to cause instabilities.
Normally, the bonding driver is UP before any enslavement takes place.
Once a netdevice is UP, the network stack acts to have it join some multicast groups
(eg the all-hosts 224.0.0.1). Now, since ether_setup() have set the bonding device
type to be ARPHRD_ETHER and address len to be ETHER_ALEN, the net core code
computes a wrong multicast link address. This is b/c ip_eth_mc_map() is called
where for multicast joins taking place after the enslavement another ip_xxx_mc_map()
is called (eg ip_ib_mc_map() when the bond type is ARPHRD_INFINIBAND)
Signed-off-by: Moni Shoua <monis@voltaire.com>
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
---
drivers/net/bonding/bond_main.c | 4 ++--
drivers/net/bonding/bond_sysfs.c | 6 ++----
2 files changed, 4 insertions(+), 6 deletions(-)
Index: net-2.6/drivers/net/bonding/bond_main.c
===================================================================
--- net-2.6.orig/drivers/net/bonding/bond_main.c 2007-07-25 15:04:50.000000000 +0300
+++ net-2.6/drivers/net/bonding/bond_main.c 2007-07-25 15:06:17.175820632 +0300
@@ -1325,8 +1325,8 @@ int bond_enslave(struct net_device *bond
/* bond must be initialized by bond_open() before enslaving */
if (!(bond_dev->flags & IFF_UP)) {
- dprintk("Error, master_dev is not up\n");
- return -EPERM;
+ printk(KERN_WARNING DRV_NAME
+ " %s: master_dev is not up in bond_enslave\n", bond_dev->name);
}
/* already enslaved */
Index: net-2.6/drivers/net/bonding/bond_sysfs.c
===================================================================
--- net-2.6.orig/drivers/net/bonding/bond_sysfs.c 2007-07-25 14:18:12.000000000 +0300
+++ net-2.6/drivers/net/bonding/bond_sysfs.c 2007-07-25 15:06:17.176820452 +0300
@@ -266,11 +266,9 @@ static ssize_t bonding_store_slaves(stru
/* Quick sanity check -- is the bond interface up? */
if (!(bond->dev->flags & IFF_UP)) {
- printk(KERN_ERR DRV_NAME
- ": %s: Unable to update slaves because interface is down.\n",
+ printk(KERN_WARNING DRV_NAME
+ ": %s: doing slave updates when interface is down.\n",
bond->dev->name);
- ret = -EPERM;
- goto out;
}
/* Note: We can't hold bond->lock here, as bond_create grabs it. */
^ permalink raw reply
* [ofa-general] [PATCH V3 6/7] net/bonding: Handlle wrong assumptions that slave is always an Ethernet device
From: Moni Shoua @ 2007-07-30 12:54 UTC (permalink / raw)
To: rdreier, davem, fubar; +Cc: netdev, general
In-Reply-To: <46ADDB89.5030601@voltaire.com>
bonding sometimes uses Ethernet constants (such as MTU and address length) which
are not good when it enslaves non Ethernet devices (such as InfiniBand).
Signed-off-by: Moni Shoua <monis@voltaire.com>
---
drivers/net/bonding/bond_main.c | 2 +-
drivers/net/bonding/bond_sysfs.c | 10 ++++++++--
drivers/net/bonding/bonding.h | 1 +
3 files changed, 10 insertions(+), 3 deletions(-)
Index: net-2.6/drivers/net/bonding/bond_main.c
===================================================================
--- net-2.6.orig/drivers/net/bonding/bond_main.c 2007-07-25 15:06:17.000000000 +0300
+++ net-2.6/drivers/net/bonding/bond_main.c 2007-07-25 15:33:25.012883360 +0300
@@ -1255,7 +1255,7 @@ static int bond_compute_features(struct
unsigned long features = BOND_INTERSECT_FEATURES;
struct slave *slave;
struct net_device *bond_dev = bond->dev;
- unsigned short max_hard_header_len = ETH_HLEN;
+ u16 max_hard_header_len = max((u16)ETH_HLEN, bond_dev->hard_header_len);
int i;
bond_for_each_slave(bond, slave, i) {
Index: net-2.6/drivers/net/bonding/bond_sysfs.c
===================================================================
--- net-2.6.orig/drivers/net/bonding/bond_sysfs.c 2007-07-25 15:06:17.000000000 +0300
+++ net-2.6/drivers/net/bonding/bond_sysfs.c 2007-07-25 15:20:10.224527636 +0300
@@ -260,6 +260,7 @@ static ssize_t bonding_store_slaves(stru
char command[IFNAMSIZ + 1] = { 0, };
char *ifname;
int i, res, found, ret = count;
+ u32 original_mtu;
struct slave *slave;
struct net_device *dev = NULL;
struct bonding *bond = to_bond(d);
@@ -325,6 +326,7 @@ static ssize_t bonding_store_slaves(stru
}
/* Set the slave's MTU to match the bond */
+ original_mtu = dev->mtu;
if (dev->mtu != bond->dev->mtu) {
if (dev->change_mtu) {
res = dev->change_mtu(dev,
@@ -339,6 +341,9 @@ static ssize_t bonding_store_slaves(stru
}
rtnl_lock();
res = bond_enslave(bond->dev, dev);
+ bond_for_each_slave(bond, slave, i)
+ if (strnicmp(slave->dev->name, ifname, IFNAMSIZ) == 0)
+ slave->original_mtu=original_mtu;
rtnl_unlock();
if (res) {
ret = res;
@@ -351,6 +356,7 @@ static ssize_t bonding_store_slaves(stru
bond_for_each_slave(bond, slave, i)
if (strnicmp(slave->dev->name, ifname, IFNAMSIZ) == 0) {
dev = slave->dev;
+ original_mtu = slave->original_mtu;
break;
}
if (dev) {
@@ -365,9 +371,9 @@ static ssize_t bonding_store_slaves(stru
}
/* set the slave MTU to the default */
if (dev->change_mtu) {
- dev->change_mtu(dev, 1500);
+ dev->change_mtu(dev, original_mtu);
} else {
- dev->mtu = 1500;
+ dev->mtu = original_mtu;
}
}
else {
Index: net-2.6/drivers/net/bonding/bonding.h
===================================================================
--- net-2.6.orig/drivers/net/bonding/bonding.h 2007-07-25 15:03:32.000000000 +0300
+++ net-2.6/drivers/net/bonding/bonding.h 2007-07-25 15:20:10.223527810 +0300
@@ -156,6 +156,7 @@ struct slave {
s8 link; /* one of BOND_LINK_XXXX */
s8 state; /* one of BOND_STATE_XXXX */
u32 original_flags;
+ u32 original_mtu;
u32 link_failure_count;
u16 speed;
u8 duplex;
^ permalink raw reply
* [ofa-general] [PATCH V3 7/7] net/bonding: Delay sending of gratuitous ARP to avoid failure
From: Moni Shoua @ 2007-07-30 12:56 UTC (permalink / raw)
To: rdreier, davem, fubar; +Cc: netdev, general
In-Reply-To: <46ADDB89.5030601@voltaire.com>
Delay sending a gratuitous_arp when LINK_STATE_LINKWATCH_PENDING bit
in dev->state field is on. This improves the chances for the arp packet to
be transmitted.
Signed-off-by: Moni Shoua <monis@voltaire.com>
---
drivers/net/bonding/bond_main.c | 25 +++++++++++++++++++++----
drivers/net/bonding/bonding.h | 1 +
2 files changed, 22 insertions(+), 4 deletions(-)
Index: net-2.6/drivers/net/bonding/bond_main.c
===================================================================
--- net-2.6.orig/drivers/net/bonding/bond_main.c 2007-07-25 15:33:25.000000000 +0300
+++ net-2.6/drivers/net/bonding/bond_main.c 2007-07-26 18:42:59.296296622 +0300
@@ -1134,8 +1134,13 @@ void bond_change_active_slave(struct bon
if (new_active && !bond->do_set_mac_addr)
memcpy(bond->dev->dev_addr, new_active->dev->dev_addr,
new_active->dev->addr_len);
-
- bond_send_gratuitous_arp(bond);
+ if (bond->curr_active_slave &&
+ test_bit(__LINK_STATE_LINKWATCH_PENDING, &bond->curr_active_slave->dev->state)){
+ dprintk("delaying gratuitous arp on %s\n",bond->curr_active_slave->dev->name);
+ bond->send_grat_arp=1;
+ }else{
+ bond_send_gratuitous_arp(bond);
+ }
}
}
@@ -2120,6 +2125,15 @@ void bond_mii_monitor(struct net_device
* program could monitor the link itself if needed.
*/
+ if (bond->send_grat_arp) {
+ if (bond->curr_active_slave && test_bit(__LINK_STATE_LINKWATCH_PENDING, &bond->curr_active_slave->dev->state))
+ dprintk("Needs to send gratuitous arp but not yet\n",__FUNCTION__);
+ else {
+ dprintk("sending delayed gratuitous arp on ond->curr_active_slave->dev->name\n");
+ bond_send_gratuitous_arp(bond);
+ bond->send_grat_arp=0;
+ }
+ }
read_lock(&bond->curr_slave_lock);
oldcurrent = bond->curr_active_slave;
read_unlock(&bond->curr_slave_lock);
@@ -2513,6 +2527,7 @@ static void bond_send_gratuitous_arp(str
struct slave *slave = bond->curr_active_slave;
struct vlan_entry *vlan;
struct net_device *vlan_dev;
+ int i;
dprintk("bond_send_grat_arp: bond %s slave %s\n", bond->dev->name,
slave ? slave->dev->name : "NULL");
@@ -2520,8 +2535,9 @@ static void bond_send_gratuitous_arp(str
return;
if (bond->master_ip) {
- bond_arp_send(slave->dev, ARPOP_REPLY, bond->master_ip,
- bond->master_ip, 0);
+ for (i=0;i<3;i++)
+ bond_arp_send(slave->dev, ARPOP_REPLY, bond->master_ip,
+ bond->master_ip, 0);
}
list_for_each_entry(vlan, &bond->vlan_list, vlan_list) {
@@ -4331,6 +4347,7 @@ static int bond_init(struct net_device *
bond->current_arp_slave = NULL;
bond->primary_slave = NULL;
bond->dev = bond_dev;
+ bond->send_grat_arp=0;
INIT_LIST_HEAD(&bond->vlan_list);
/* Initialize the device entry points */
Index: net-2.6/drivers/net/bonding/bonding.h
===================================================================
--- net-2.6.orig/drivers/net/bonding/bonding.h 2007-07-25 15:20:10.000000000 +0300
+++ net-2.6/drivers/net/bonding/bonding.h 2007-07-26 18:42:43.652087660 +0300
@@ -203,6 +203,7 @@ struct bonding {
struct vlan_group *vlgrp;
struct packet_type arp_mon_pt;
s8 do_set_mac_addr;
+ int send_grat_arp;
};
/**
^ permalink raw reply
* Re: 2.6.20->2.6.21 - networking dies after random time
From: Alan Cox @ 2007-07-30 13:05 UTC (permalink / raw)
To: Ingo Molnar
Cc: Jarek Poplawski, Thomas Gleixner, Linus Torvalds, Marcin ??lusarz,
Jean-Baptiste Vignaud, linux-kernel, shemminger, linux-net,
netdev, Andrew Morton
In-Reply-To: <20070730084651.GA14146@elte.hu>
> So the whole locking is to be able to keep irqs enabled for a long time,
> without risking entry of the same IRQ handler on this same CPU, correct?
As implemented - on any CPU.
We also need to know that the IRQ handler is not doing useful work on
another processor which is why we take the lock after disabling the
interrupt line everywhere. Without that we might be completing an IRQ on
another CPU and that would race the transmit and make a nasty mess.
> So it seems to me that maybe the driver could be surprised via these
> spurious interrupts that happen right after the irq_enable(). Does the
> patch below make any sense in your opinion?
For MMIO it does look like that may be needed. Looks sensible.
^ permalink raw reply
* Re: [PATCH 1/7] Preparatory refactoring part 1.
From: Patrick McHardy @ 2007-07-30 13:51 UTC (permalink / raw)
To: Corey Hickey; +Cc: netdev
In-Reply-To: <11857548774008-git-send-email-bugfood-ml@fatooh.org>
Corey Hickey wrote:
> diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
> index 9579573..8ae077f 100644
> --- a/net/sched/sch_sfq.c
> +++ b/net/sched/sch_sfq.c
> @@ -77,6 +77,9 @@
> #define SFQ_DEPTH 128
> #define SFQ_HASH_DIVISOR 1024
>
> +#define SFQ_HEAD 0
> +#define SFQ_TAIL 1
> +
> /* This type should contain at least SFQ_DEPTH*2 values */
> typedef unsigned char sfq_index;
>
> @@ -244,10 +247,8 @@ static unsigned int sfq_drop(struct Qdisc *sch)
> return 0;
> }
>
> -static int
> -sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
> +static void sfq_q_enqueue(struct sk_buff *skb, struct sfq_sched_data *q, unsigned int end)
Please make sure to break at 80 chars and to keep the style
in this file consistent (newline before function name).
> {
> - struct sfq_sched_data *q = qdisc_priv(sch);
> unsigned hash = sfq_hash(q, skb);
> sfq_index x;
>
> @@ -256,8 +257,12 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
> q->ht[hash] = x = q->dep[SFQ_DEPTH].next;
> q->hash[x] = hash;
> }
> - sch->qstats.backlog += skb->len;
Why not keep this instead of having both callers do it?
> - __skb_queue_tail(&q->qs[x], skb);
> +
> + if (end == SFQ_TAIL)
> + __skb_queue_tail(&q->qs[x], skb);
> + else
> + __skb_queue_head(&q->qs[x], skb);
> +
> sfq_inc(q, x);
> if (q->qs[x].qlen == 1) { /* The flow is new */
> if (q->tail == SFQ_DEPTH) { /* It is the first flow */
> @@ -270,12 +275,21 @@ sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
> q->tail = x;
> }
> }
> +}
> +
> +static int
> +sfq_enqueue(struct sk_buff *skb, struct Qdisc* sch)
> +{
> + struct sfq_sched_data *q = qdisc_priv(sch);
newline please.
> + sfq_q_enqueue(skb, q, SFQ_TAIL);
> + sch->qstats.backlog += skb->len;
> if (++sch->q.qlen < q->limit-1) {
> sch->bstats.bytes += skb->len;
> sch->bstats.packets++;
> return 0;
> }
>
> + sch->qstats.drops++;
sfq_drop already increments this.
> sfq_drop(sch);
> return NET_XMIT_CN;
> }
> @@ -284,28 +298,8 @@ static int
> sfq_requeue(struct sk_buff *skb, struct Qdisc* sch)
> {
> struct sfq_sched_data *q = qdisc_priv(sch);
newline please
> - unsigned hash = sfq_hash(q, skb);
> - sfq_index x;
> -
> - x = q->ht[hash];
> - if (x == SFQ_DEPTH) {
> - q->ht[hash] = x = q->dep[SFQ_DEPTH].next;
> - q->hash[x] = hash;
> - }
> + sfq_q_enqueue(skb, q, SFQ_HEAD);
> sch->qstats.backlog += skb->len;
> - __skb_queue_head(&q->qs[x], skb);
> - sfq_inc(q, x);
> - if (q->qs[x].qlen == 1) { /* The flow is new */
> - if (q->tail == SFQ_DEPTH) { /* It is the first flow */
> - q->tail = x;
> - q->next[x] = x;
> - q->allot[x] = q->quantum;
> - } else {
> - q->next[x] = q->next[q->tail];
> - q->next[q->tail] = x;
> - q->tail = x;
> - }
> - }
> if (++sch->q.qlen < q->limit - 1) {
> sch->qstats.requeues++;
> return 0;
> @@ -316,13 +310,8 @@ sfq_requeue(struct sk_buff *skb, struct Qdisc* sch)
> return NET_XMIT_CN;
> }
>
> -
> -
> -
> -static struct sk_buff *
> -sfq_dequeue(struct Qdisc* sch)
> +static struct sk_buff *sfq_q_dequeue(struct sfq_sched_data *q)
Keep style consistent please.
> {
> - struct sfq_sched_data *q = qdisc_priv(sch);
> struct sk_buff *skb;
> sfq_index a, old_a;
>
> @@ -335,8 +324,6 @@ sfq_dequeue(struct Qdisc* sch)
> /* Grab packet */
> skb = __skb_dequeue(&q->qs[a]);
> sfq_dec(q, a);
> - sch->q.qlen--;
> - sch->qstats.backlog -= skb->len;
>
> /* Is the slot empty? */
> if (q->qs[a].qlen == 0) {
> @@ -353,6 +340,21 @@ sfq_dequeue(struct Qdisc* sch)
> a = q->next[a];
> q->allot[a] += q->quantum;
> }
> +
> + return skb;
> +}
> +
> +static struct sk_buff
> +*sfq_dequeue(struct Qdisc* sch)
> +{
> + struct sfq_sched_data *q = qdisc_priv(sch);
> + struct sk_buff *skb;
> +
> + skb = sfq_q_dequeue(q);
> + if (skb == NULL)
> + return NULL;
> + sch->q.qlen--;
> + sch->qstats.backlog -= skb->len;
> return skb;
> }
>
^ permalink raw reply
* Re: [PATCH 2/7] Preparatory refactoring part 2.
From: Patrick McHardy @ 2007-07-30 13:59 UTC (permalink / raw)
To: Corey Hickey; +Cc: netdev
In-Reply-To: <11857548771609-git-send-email-bugfood-ml@fatooh.org>
Corey Hickey wrote:
> diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
> index 8ae077f..0c46938 100644
> --- a/net/sched/sch_sfq.c
> +++ b/net/sched/sch_sfq.c
> @@ -380,71 +380,71 @@ static void sfq_perturbation(unsigned long arg)
> }
> }
>
> -static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
> +static int sfq_q_init(struct sfq_sched_data *q, struct rtattr *opt)
> {
> - struct sfq_sched_data *q = qdisc_priv(sch);
> struct tc_sfq_qopt *ctl = RTA_DATA(opt);
> - unsigned int qlen;
> + int i;
>
> - if (opt->rta_len < RTA_LENGTH(sizeof(*ctl)))
> + if (opt && opt->rta_len < RTA_LENGTH(sizeof(*ctl)))
opt is dereferenced above (RTA_DATA), so if it is NULL we've already
crashed.
> return -EINVAL;
>
> - sch_tree_lock(sch);
> - q->quantum = ctl->quantum ? : psched_mtu(sch->dev);
> - q->perturb_period = ctl->perturb_period*HZ;
> - if (ctl->limit)
> - q->limit = min_t(u32, ctl->limit, SFQ_DEPTH);
> + q->perturbation = 0;
> + q->max_depth = 0;
> + q->tail = q->limit = SFQ_DEPTH;
> + if (opt == NULL) {
> + q->perturb_period = 0;
> + } else {
> + struct tc_sfq_qopt *ctl = RTA_DATA(opt);
> + if (ctl->quantum)
> + q->quantum = ctl->quantum;
> + q->perturb_period = ctl->perturb_period*HZ;
>
> - qlen = sch->q.qlen;
> - while (sch->q.qlen >= q->limit-1)
> - sfq_drop(sch);
> - qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
I hope that patch that makes changing possible brings this back ..
<checking> .. it doesn't. Please either keep this or fix up 6/7
to bring it back.
> + if (ctl->limit)
> + q->limit = min_t(u32, ctl->limit, SFQ_DEPTH);
> + }
>
> - del_timer(&q->perturb_timer);
> - if (q->perturb_period) {
> - q->perturb_timer.expires = jiffies + q->perturb_period;
> - add_timer(&q->perturb_timer);
> + for (i=0; i<SFQ_HASH_DIVISOR; i++)
> + q->ht[i] = SFQ_DEPTH;
> + for (i=0; i<SFQ_DEPTH; i++) {
> + skb_queue_head_init(&q->qs[i]);
> + q->dep[i+SFQ_DEPTH].next = i+SFQ_DEPTH;
> + q->dep[i+SFQ_DEPTH].prev = i+SFQ_DEPTH;
> }
> - sch_tree_unlock(sch);
> +
> + for (i=0; i<SFQ_DEPTH; i++)
> + sfq_link(q, i);
> return 0;
> }
>
> +static void sfq_q_destroy(struct sfq_sched_data *q)
> +{
> + del_timer(&q->perturb_timer);
> +}
> +
> static void sfq_destroy(struct Qdisc *sch)
> {
> struct sfq_sched_data *q = qdisc_priv(sch);
> - del_timer(&q->perturb_timer);
> + sfq_q_destroy(q);
> }
That really does look a bit pointless.
^ permalink raw reply
* Re: [PATCH 6/7] Make qdisc changeable.
From: Patrick McHardy @ 2007-07-30 14:11 UTC (permalink / raw)
To: Corey Hickey; +Cc: netdev
In-Reply-To: <11857548783311-git-send-email-bugfood-ml@fatooh.org>
Corey Hickey wrote:
> Re-implement sfq_change() and enable Qdisc_opts.change so "tc qdisc
> change" will work.
>
> Signed-off-by: Corey Hickey <bugfood-ml@fatooh.org>
> ---
> net/sched/sch_sfq.c | 51 ++++++++++++++++++++++++++++++++++++++++++++++++++-
> 1 files changed, 50 insertions(+), 1 deletions(-)
>
> diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
> index e6a6a21..e042cd0 100644
> --- a/net/sched/sch_sfq.c
> +++ b/net/sched/sch_sfq.c
> @@ -485,6 +485,55 @@ static int sfq_init(struct Qdisc *sch, struct rtattr *opt)
> return 0;
> }
>
> +static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
> +{
> + struct sfq_sched_data *q = qdisc_priv(sch);
> + struct sfq_sched_data tmp;
> + struct sk_buff *skb;
> + int err;
> +
> + /* set up tmp queue */
> + memset(&tmp, 0, sizeof(struct sfq_sched_data));
> + tmp.quantum = psched_mtu(sch->dev); /* default */
If no value is given it should use the old value instead of
reinitializing to the default.
> + if ((err = sfq_q_init(&tmp, opt)))
> + return err;
This will also use defaults for all unspecified values. It would
be more consistent with other qdiscs to only change those values
that are actually specified, so something like "tc qdisc change ...
perturb 10" will *only* change the perturbation parameter.
I'm not sure reusing the initialization function and copying the
parameters is the best way to do this.
> +
> + /* copy packets from the old queue to the tmp queue */
> + sch_tree_lock(sch);
> + while (sch->q.qlen >= tmp.limit - 1)
> + sfq_drop(sch);
> + while ((skb = sfq_q_dequeue(q)) != NULL)
> + sfq_q_enqueue(skb, &tmp, SFQ_TAIL);
> +
> + /* clean up the old queue */
> + sfq_q_destroy(q);
> +
> + /* copy elements of the tmp queue into the old queue */
> + q->perturb_period = tmp.perturb_period;
> + q->quantum = tmp.quantum;
> + q->limit = tmp.limit;
> + q->depth = tmp.depth;
> + q->hash_divisor = tmp.hash_divisor;
> + q->tail = tmp.tail;
> + q->max_depth = tmp.max_depth;
> + q->ht = tmp.ht;
> + q->dep = tmp.dep;
> + q->next = tmp.next;
> + q->allot = tmp.allot;
> + q->hash = tmp.hash;
> + q->qs = tmp.qs;
> +
> + /* finish up */
> + if (q->perturb_period) {
> + q->perturb_timer.expires = jiffies + q->perturb_period;
> + add_timer(&q->perturb_timer);
> + } else {
> + q->perturbation = 0;
> + }
> + sch_tree_unlock(sch);
> + return 0;
> +}
> +
^ permalink raw reply
* Re: [PATCH] NET_DMA: remove unused dma_memcpy_to_kernel_iovec
From: pravin @ 2007-07-30 14:39 UTC (permalink / raw)
To: David Miller; +Cc: shannon.nelson, netdev, viro, christopher.leech, andy.grover
In-Reply-To: <20070726.000607.18303016.davem@davemloft.net>
On 7/26/07, David Miller <davem@davemloft.net> wrote:
> From: Shannon Nelson <shannon.nelson@intel.com>
> Date: Tue, 24 Jul 2007 17:36:06 -0700
>
> > (repost - original eaten by vger?)
> >
> > Al Viro pointed out that dma_memcpy_to_kernel_iovec() really was
> > unreachable and thus unused. The code originally was there to support
> > in-kernel dma needs, but since it remains unused, we'll pull it out.
> >
> > Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
>
> Applied, thanks Shannon.
NET_DMA on kernel buffer is pretty useful in ndb, iSCSI target and
initiators which uses kernel buffer to receive data. Are there any
other issues with dma-memcpy on kernel buffers, if not then following
patch makes dma_memcpy_to_kernel_iovec() reachable from tcp_recvmsg.
I tested this patch and it work fine with unh iSCSI target.
comments?
--pravin.
Index: linux-2.6.23-rc1/net/ipv4/tcp.c
===================================================================
--- linux-2.6.23-rc1.orig/net/ipv4/tcp.c 2007-07-23 02:11:00.000000000 +0530
+++ linux-2.6.23-rc1/net/ipv4/tcp.c 2007-07-30 17:43:51.000000000 +0530
@@ -1115,7 +1115,7 @@
int target; /* Read at least this many bytes */
long timeo;
struct task_struct *user_recv = NULL;
- int copied_early = 0;
+ int copied_early = 0, kernel_dma = 0;
struct sk_buff *skb;
lock_sock(sk);
@@ -1154,6 +1154,9 @@
!sysctl_tcp_low_latency &&
__get_cpu_var(softnet_data).net_dma) {
preempt_enable_no_resched();
+ if (segment_eq(get_fs(), KERNEL_DS))
+ kernel_dma = 1;
+ else
tp->ucopy.pinned_list =
dma_pin_iovec_pages(msg->msg_iov, len);
} else {
@@ -1362,7 +1365,7 @@
if (!(flags & MSG_TRUNC)) {
#ifdef CONFIG_NET_DMA
- if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
+ if (!tp->ucopy.dma_chan && (kernel_dma || tp->ucopy.pinned_list))
tp->ucopy.dma_chan = get_softnet_dma();
if (tp->ucopy.dma_chan) {
> -
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply
* Re: [PATCH RFC]: napi_struct V4
From: Roland Dreier @ 2007-07-30 15:04 UTC (permalink / raw)
To: David Miller; +Cc: netdev, shemminger, jgarzik, hadi, rusty
In-Reply-To: <20070728.223206.112621083.davem@davemloft.net>
> If you have a means in the device (like tg3, bnx2, e1000, and a score
> of others do) to force the device to trigger a HW interrupt, that's
> what you do if you detect that events are pending after re-enabling
> interrupt in the ->poll() handler.
It is possible to trigger an interrupt for IPoIB I think but it will
lead to a fair bit of code, since one would have to do something like
a dummy send operation to generate an event that leads to the
interrupt. I think this will be rather complex/ugly to implement
because we have to make sure we have space in the send queue to post
the send operation, etc.
> Frankly I don't think the lock is a big deal and you need something
> like it anyways typically.
If I understand this correctly, you're suggesting a spin_lock_irqsave()
around the netif_rx_complete() in the poll routine, and a
corresponding lock in the interrupt handler. That seems like a pretty
big step backwards for performance to me. Especially since in my
experience, fast machines handling full-MTU traffic often end up being
basically interrupt driven because they drain the RX ring too quickly
to stay in NAPI polling. Yes, it's "only one more lock" but look at
the tricky smp_mb() usage that tg3, bnx2, etc have to avoid using a
spinlock...
IPoIB can cope but it really seems like an unfortunate feature of
these changes that we can't do something like what we have today,
which imposes no overhead unless an event actually lands in the race
window.
- R.
^ permalink raw reply
* [PATCH 1/3] ehea: Fix workqueue handling
From: Thomas Klein @ 2007-07-30 15:50 UTC (permalink / raw)
To: Jeff Garzik
Cc: Thomas Klein, Jan-Bernd Themann, netdev, linux-kernel,
Christoph Raisch, Stefan Roscher, linux-ppc, Jan-Bernd Themann,
Marcus Eder
Fix: Workqueue ehea_driver_wq was not destroyed
Signed-off-by: Thomas Klein <tklein@de.ibm.com>
---
drivers/net/ehea/ehea.h | 2 +-
drivers/net/ehea/ehea_main.c | 1 +
2 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h
index 8ee2c2c..d67f97b 100644
--- a/drivers/net/ehea/ehea.h
+++ b/drivers/net/ehea/ehea.h
@@ -39,7 +39,7 @@
#include <asm/io.h>
#define DRV_NAME "ehea"
-#define DRV_VERSION "EHEA_0072"
+#define DRV_VERSION "EHEA_0073"
/* eHEA capability flags */
#define DLPAR_PORT_ADD_REM 1
diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
index 58702f5..d43ab0f 100644
--- a/drivers/net/ehea/ehea_main.c
+++ b/drivers/net/ehea/ehea_main.c
@@ -3099,6 +3099,7 @@ out:
static void __exit ehea_module_exit(void)
{
+ destroy_workqueue(ehea_driver_wq);
driver_remove_file(&ehea_driver.driver, &driver_attr_capabilities);
ibmebus_unregister_driver(&ehea_driver);
ehea_destroy_busmap();
--
1.5.2
^ permalink raw reply related
* [PATCH 2/3] ehea: Simplify resource usage check
From: Thomas Klein @ 2007-07-30 15:50 UTC (permalink / raw)
To: Jeff Garzik
Cc: Christoph Raisch, Jan-Bernd Themann, Jan-Bernd Themann,
linux-kernel, linux-ppc, Marcus Eder, netdev, Thomas Klein,
Stefan Roscher
Use shorter method to determine whether adapter has configured ports
Signed-off-by: Thomas Klein <tklein@de.ibm.com>
---
drivers/net/ehea/ehea_main.c | 18 ++++++------------
1 files changed, 6 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
index d43ab0f..36ca322 100644
--- a/drivers/net/ehea/ehea_main.c
+++ b/drivers/net/ehea/ehea_main.c
@@ -2165,24 +2165,18 @@ static int ehea_clean_all_portres(struct ehea_port *port)
return ret;
}
-static void ehea_remove_adapter_mr (struct ehea_adapter *adapter)
+static void ehea_remove_adapter_mr(struct ehea_adapter *adapter)
{
- int i;
-
- for (i=0; i < EHEA_MAX_PORTS; i++)
- if (adapter->port[i])
- return;
+ if (adapter->active_ports)
+ return;
ehea_rem_mr(&adapter->mr);
}
-static int ehea_add_adapter_mr (struct ehea_adapter *adapter)
+static int ehea_add_adapter_mr(struct ehea_adapter *adapter)
{
- int i;
-
- for (i=0; i < EHEA_MAX_PORTS; i++)
- if (adapter->port[i])
- return 0;
+ if (adapter->active_ports)
+ return 0;
return ehea_reg_kernel_mr(adapter, &adapter->mr);
}
--
1.5.2
^ permalink raw reply related
* [PATCH 3/3] ehea: Eliminated some compiler warnings
From: Thomas Klein @ 2007-07-30 15:52 UTC (permalink / raw)
To: Jeff Garzik
Cc: Christoph Raisch, Jan-Bernd Themann, Jan-Bernd Themann,
linux-kernel, linux-ppc, Marcus Eder, netdev, Thomas Klein,
Stefan Roscher
Fixed wrongly casted pointers
Signed-off-by: Thomas Klein <tklein@de.ibm.com>
---
drivers/net/ehea/ehea_main.c | 25 ++++++++++---------------
1 files changed, 10 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
index 36ca322..9756211 100644
--- a/drivers/net/ehea/ehea_main.c
+++ b/drivers/net/ehea/ehea_main.c
@@ -1326,7 +1326,6 @@ static void write_swqe2_TSO(struct sk_buff *skb,
u8 *imm_data = &swqe->u.immdata_desc.immediate_data[0];
int skb_data_size = skb->len - skb->data_len;
int headersize;
- u64 tmp_addr;
/* Packet is TCP with TSO enabled */
swqe->tx_control |= EHEA_SWQE_TSO;
@@ -1347,9 +1346,8 @@ static void write_swqe2_TSO(struct sk_buff *skb,
/* set sg1entry data */
sg1entry->l_key = lkey;
sg1entry->len = skb_data_size - headersize;
-
- tmp_addr = (u64)(skb->data + headersize);
- sg1entry->vaddr = ehea_map_vaddr(tmp_addr);
+ sg1entry->vaddr =
+ ehea_map_vaddr(skb->data + headersize);
swqe->descriptors++;
}
} else
@@ -1362,7 +1360,6 @@ static void write_swqe2_nonTSO(struct sk_buff *skb,
int skb_data_size = skb->len - skb->data_len;
u8 *imm_data = &swqe->u.immdata_desc.immediate_data[0];
struct ehea_vsgentry *sg1entry = &swqe->u.immdata_desc.sg_entry;
- u64 tmp_addr;
/* Packet is any nonTSO type
*
@@ -1379,8 +1376,8 @@ static void write_swqe2_nonTSO(struct sk_buff *skb,
/* copy sg1entry data */
sg1entry->l_key = lkey;
sg1entry->len = skb_data_size - SWQE2_MAX_IMM;
- tmp_addr = (u64)(skb->data + SWQE2_MAX_IMM);
- sg1entry->vaddr = ehea_map_vaddr(tmp_addr);
+ sg1entry->vaddr =
+ ehea_map_vaddr(skb->data + SWQE2_MAX_IMM);
swqe->descriptors++;
}
} else {
@@ -1395,7 +1392,6 @@ static inline void write_swqe2_data(struct sk_buff *skb, struct net_device *dev,
struct ehea_vsgentry *sg_list, *sg1entry, *sgentry;
skb_frag_t *frag;
int nfrags, sg1entry_contains_frag_data, i;
- u64 tmp_addr;
nfrags = skb_shinfo(skb)->nr_frags;
sg1entry = &swqe->u.immdata_desc.sg_entry;
@@ -1417,9 +1413,9 @@ static inline void write_swqe2_data(struct sk_buff *skb, struct net_device *dev,
/* copy sg1entry data */
sg1entry->l_key = lkey;
sg1entry->len = frag->size;
- tmp_addr = (u64)(page_address(frag->page)
- + frag->page_offset);
- sg1entry->vaddr = ehea_map_vaddr(tmp_addr);
+ sg1entry->vaddr =
+ ehea_map_vaddr(page_address(frag->page)
+ + frag->page_offset);
swqe->descriptors++;
sg1entry_contains_frag_data = 1;
}
@@ -1431,10 +1427,9 @@ static inline void write_swqe2_data(struct sk_buff *skb, struct net_device *dev,
sgentry->l_key = lkey;
sgentry->len = frag->size;
-
- tmp_addr = (u64)(page_address(frag->page)
- + frag->page_offset);
- sgentry->vaddr = ehea_map_vaddr(tmp_addr);
+ sgentry->vaddr =
+ ehea_map_vaddr(page_address(frag->page)
+ + frag->page_offset);
swqe->descriptors++;
}
}
--
1.5.2
^ permalink raw reply related
* [PATCH 1/4][RFC] lro: Generic Large Receive Offload for TCP traffic
From: Jan-Bernd Themann @ 2007-07-30 15:24 UTC (permalink / raw)
To: netdev
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, Stefan Roscher, David Miller,
Andrew Gallatin, Jeff Garzik
Generic Large Receive Offload for TCP traffic
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
---
include/linux/inet_lro.h | 173 ++++++++++++++
net/ipv4/inet_lro.c | 590 ++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 763 insertions(+), 0 deletions(-)
create mode 100644 include/linux/inet_lro.h
create mode 100644 net/ipv4/inet_lro.c
diff --git a/include/linux/inet_lro.h b/include/linux/inet_lro.h
new file mode 100644
index 0000000..0957234
--- /dev/null
+++ b/include/linux/inet_lro.h
@@ -0,0 +1,173 @@
+/*
+ * linux/include/linux/inet_lro.h
+ *
+ * Large Receive Offload (ipv4 / tcp)
+ *
+ * (C) Copyright IBM Corp. 2007
+ *
+ * Authors:
+ * Jan-Bernd Themann <themann@de.ibm.com>
+ * Christoph Raisch <raisch@de.ibm.com>
+ *
+ *
+ * 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, 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef __INET_LRO_H_
+#define __INET_LRO_H_
+
+#include <net/ip.h>
+#include <net/tcp.h>
+
+/*
+ * LRO statistics
+ */
+
+struct net_lro_stats {
+ unsigned long aggregated;
+ unsigned long flushed;
+ unsigned long no_desc;
+};
+
+/*
+ * LRO descriptor for a tcp session
+ */
+struct net_lro_desc {
+ struct sk_buff *parent;
+ struct sk_buff *last_skb;
+ struct skb_frag_struct *next_frag;
+ struct iphdr *iph;
+ struct tcphdr *tcph;
+ struct vlan_group *vgrp;
+ __wsum data_csum;
+ u32 tcp_rcv_tsecr;
+ u32 tcp_rcv_tsval;
+ u32 tcp_ack;
+ u32 tcp_next_seq;
+ u32 skb_tot_frags_len;
+ u16 ip_tot_len;
+ u16 tcp_saw_tstamp; /* timestamps enabled */
+ u16 tcp_window;
+ u16 vlan_tag;
+ int pkt_aggr_cnt; /* counts aggregated packets */
+ int vlan_packet;
+ int active;
+};
+
+/*
+ * Large Receive Offload (LRO) Manager
+ *
+ * Fields must be set by driver
+ */
+
+struct net_lro_mgr {
+ struct net_device *dev;
+ struct net_lro_stats stats;
+
+ /* LRO features */
+ unsigned long features;
+#define LRO_F_NAPI 1 /* Pass packets to stack via NAPI */
+#define LRO_F_EXTRACT_VLAN_ID 2 /* Set flag if VLAN IDs are extracted
+ from received packets and eth protocol
+ is still ETH_P_8021Q */
+
+ u32 ip_summed; /* Options to be set in generated SKB in page mode */
+ int max_desc; /* Max number of LRO descriptors */
+ int max_aggr; /* Max number of LRO packets to be aggregated */
+
+ struct net_lro_desc *lro_arr; /* Array of LRO descriptors */
+
+ /*
+ * Optimized driver functions
+ *
+ * get_skb_header: returns tcp and ip header for packet in SKB
+ */
+ int (*get_skb_header)(struct sk_buff *skb, void **ip_hdr,
+ void **tcpudp_hdr, u64 *hdr_flags, void *priv);
+
+ /* hdr_flags: */
+#define LRO_IPV4 1 /* ip_hdr is IPv4 header */
+#define LRO_TCP 2 /* tcpudp_hdr is TCP header */
+
+ /*
+ * get_frag_header: returns mac, tcp and ip header for packet in SKB
+ *
+ * @hdr_flags: Indicate what kind of LRO has to be done
+ * (IPv4/IPv6/TCP/UDP)
+ */
+ int (*get_frag_header)(struct skb_frag_struct *frag, void **mac_hdr,
+ void **ip_hdr, void **tcpudp_hdr, u64 *hdr_flags,
+ void *priv);
+};
+
+/*
+ * Processes a SKB
+ *
+ * @lro_mgr: LRO manager to use
+ * @skb: SKB to aggregate
+ * @priv: Private data that may be used by driver functions
+ * (for example get_tcp_ip_hdr)
+ */
+
+void lro_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ void *priv);
+
+/*
+ * Processes a SKB with VLAN HW acceleration support
+ */
+
+void lro_vlan_hwaccel_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ struct vlan_group *vgrp,
+ u16 vlan_tag,
+ void *priv);
+
+/*
+ * Processes a fragment list
+ *
+ * This functions aggregate fragments and generate SKBs do pass
+ * the packets to the stack.
+ *
+ * @lro_mgr: LRO manager to use
+ * @frags: Fragment to be processed. Must contain entire header in first
+ * element.
+ * @len: Length of received data
+ * @true_size: Actual size of memory the fragment is consuming
+ * @priv: Private data that may be used by driver functions
+ * (for example get_tcp_ip_hdr)
+ */
+
+void lro_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size, void *priv, __wsum sum);
+
+void lro_vlan_hwaccel_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ struct vlan_group *vgrp,
+ u16 vlan_tag,
+ void *priv, __wsum sum);
+
+/*
+ * Forward all aggregated SKBs held by lro_mgr to network stack
+ */
+
+void lro_flush_all(struct net_lro_mgr *lro_mgr);
+
+void lro_flush_pkt(struct net_lro_mgr *lro_mgr,
+ struct iphdr *iph, struct tcphdr *tcph);
+
+#endif
diff --git a/net/ipv4/inet_lro.c b/net/ipv4/inet_lro.c
new file mode 100644
index 0000000..a01a2b8
--- /dev/null
+++ b/net/ipv4/inet_lro.c
@@ -0,0 +1,590 @@
+/*
+ * linux/net/ipv4/inet_lro.c
+ *
+ * Large Receive Offload (ipv4 / tcp)
+ *
+ * (C) Copyright IBM Corp. 2007
+ *
+ * Authors:
+ * Jan-Bernd Themann <themann@de.ibm.com>
+ * Christoph Raisch <raisch@de.ibm.com>
+ *
+ *
+ * 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, 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., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+
+#include <linux/module.h>
+#include <linux/if_vlan.h>
+#include <linux/inet_lro.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Jan-Bernd Themann <themann@de.ibm.com>");
+MODULE_DESCRIPTION("Large Receive Offload (ipv4 / tcp)");
+
+#define TCP_HDR_LEN(tcph) (tcph->doff << 2)
+#define IP_HDR_LEN(iph) (iph->ihl << 2)
+#define TCP_PAYLOAD_LENGTH(iph, tcph) \
+ (ntohs(iph->tot_len) - IP_HDR_LEN(iph) - TCP_HDR_LEN(tcph))
+
+#define IPH_LEN_WO_OPTIONS 5
+#define TCPH_LEN_WO_OPTIONS 5
+#define TCPH_LEN_W_TIMESTAMP 8
+
+#define LRO_MAX_PG_HLEN 64
+
+#define LRO_INC_STATS(lro_mgr, attr) { lro_mgr->stats.attr++; }
+
+/*
+ * Basic tcp checks whether packet is suitable for LRO
+ */
+
+static int lro_tcp_ip_check(struct iphdr *iph, struct tcphdr *tcph,
+ int len, struct net_lro_desc *lro_desc)
+{
+ /* check ip header: don't aggregate padded frames */
+ if (ntohs(iph->tot_len) != len)
+ return -1;
+
+ if (TCP_PAYLOAD_LENGTH(iph, tcph) == 0)
+ return -1;
+
+ if (iph->ihl != IPH_LEN_WO_OPTIONS)
+ return -1;
+
+ if (tcph->cwr || tcph->ece || tcph->urg || !tcph->ack
+ || tcph->rst || tcph->syn || tcph->fin)
+ return -1;
+
+ if (INET_ECN_is_ce(ipv4_get_dsfield(iph)))
+ return -1;
+
+ if (tcph->doff != TCPH_LEN_WO_OPTIONS
+ && tcph->doff != TCPH_LEN_W_TIMESTAMP)
+ return -1;
+
+ /* check tcp options (only timestamp allowed) */
+ if (tcph->doff == TCPH_LEN_W_TIMESTAMP) {
+ u32 *topt = (u32 *)(tcph + 1);
+
+ if (*topt != htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
+ | (TCPOPT_TIMESTAMP << 8)
+ | TCPOLEN_TIMESTAMP))
+ return -1;
+
+ /* timestamp should be in right order */
+ topt++;
+ if (lro_desc && after(ntohl(lro_desc->tcp_rcv_tsval),
+ ntohl(*topt)))
+ return -1;
+
+ /* timestamp reply should not be zero */
+ topt++;
+ if (*topt == 0)
+ return -1;
+ }
+
+ return 0;
+}
+
+static void lro_update_tcp_ip_header(struct net_lro_desc *lro_desc)
+{
+ struct iphdr *iph = lro_desc->iph;
+ struct tcphdr *tcph = lro_desc->tcph;
+ u32 *p;
+ __wsum tcp_hdr_csum;
+
+ tcph->ack_seq = lro_desc->tcp_ack;
+ tcph->window = lro_desc->tcp_window;
+
+ if (lro_desc->tcp_saw_tstamp) {
+ p = (u32 *)(tcph + 1);
+ *(p+2) = lro_desc->tcp_rcv_tsecr;
+ }
+
+ iph->tot_len = htons(lro_desc->ip_tot_len);
+
+ iph->check = 0;
+ iph->check = ip_fast_csum((u8 *)lro_desc->iph, iph->ihl);
+
+ tcph->check = 0;
+ tcp_hdr_csum = csum_partial((u8 *)tcph, TCP_HDR_LEN(tcph), 0);
+ lro_desc->data_csum = csum_add(lro_desc->data_csum, tcp_hdr_csum);
+ tcph->check = csum_tcpudp_magic(iph->saddr, iph->daddr,
+ htons(lro_desc->ip_tot_len) -
+ IP_HDR_LEN(iph), IPPROTO_TCP,
+ lro_desc->data_csum);
+}
+
+static __wsum lro_tcp_data_csum(struct iphdr *iph, struct tcphdr *tcph, int len)
+{
+ __wsum tcp_csum;
+ __wsum tcp_hdr_csum;
+ __wsum tcp_ps_hdr_csum;
+
+ tcp_csum = ~csum_unfold(tcph->check);
+ tcp_hdr_csum = csum_partial((u8 *)tcph, TCP_HDR_LEN(tcph), tcp_csum);
+
+ tcp_ps_hdr_csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
+ len + TCP_HDR_LEN(tcph),
+ IPPROTO_TCP, 0);
+
+ return csum_sub(csum_sub(tcp_csum, tcp_hdr_csum),
+ tcp_ps_hdr_csum);
+}
+
+static void lro_init_desc(struct net_lro_desc *lro_desc, struct sk_buff *skb,
+ struct iphdr *iph, struct tcphdr *tcph,
+ u16 vlan_tag, struct vlan_group *vgrp)
+{
+ int nr_frags;
+ u32 *ptr;
+ u32 tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
+
+ nr_frags = skb_shinfo(skb)->nr_frags;
+ lro_desc->parent = skb;
+ lro_desc->next_frag = &(skb_shinfo(skb)->frags[nr_frags]);
+ lro_desc->iph = iph;
+ lro_desc->tcph = tcph;
+ lro_desc->tcp_next_seq = ntohl(tcph->seq) + tcp_data_len;
+ lro_desc->tcp_ack = ntohl(tcph->ack_seq);
+ lro_desc->tcp_window = tcph->window;
+
+ lro_desc->pkt_aggr_cnt = 1;
+ lro_desc->ip_tot_len = ntohs(iph->tot_len);
+
+ if (tcph->doff == 8) {
+ ptr = (u32 *)(tcph+1);
+ lro_desc->tcp_saw_tstamp = 1;
+ lro_desc->tcp_rcv_tsval = *(ptr+1);
+ lro_desc->tcp_rcv_tsecr = *(ptr+2);
+ }
+
+ lro_desc->vgrp = vgrp;
+ lro_desc->vlan_tag = vlan_tag;
+ lro_desc->active = 1;
+
+ lro_desc->data_csum = lro_tcp_data_csum(iph, tcph,
+ tcp_data_len);
+}
+
+static inline void lro_clear_desc(struct net_lro_desc *lro_desc)
+{
+ memset(lro_desc, 0, sizeof(struct net_lro_desc));
+}
+
+static void lro_add_common(struct net_lro_desc *lro_desc, struct iphdr *iph,
+ struct tcphdr *tcph, int tcp_data_len)
+{
+ struct sk_buff *parent = lro_desc->parent;
+ u32 *topt;
+
+ lro_desc->pkt_aggr_cnt++;
+ lro_desc->ip_tot_len += tcp_data_len;
+ lro_desc->tcp_next_seq += tcp_data_len;
+ lro_desc->tcp_window = tcph->window;
+ lro_desc->tcp_ack = tcph->ack_seq;
+
+ /* don't update tcp_rcv_tsval, would not work with PAWS */
+ if (lro_desc->tcp_saw_tstamp) {
+ topt = (u32 *) (tcph + 1);
+ lro_desc->tcp_rcv_tsecr = *(topt + 2);
+ }
+
+ parent->len += tcp_data_len;
+ parent->data_len += tcp_data_len;
+
+ lro_desc->data_csum = csum_add(lro_desc->data_csum,
+ lro_tcp_data_csum(iph, tcph,
+ tcp_data_len));
+}
+
+static void lro_add_packet(struct net_lro_desc *lro_desc, struct sk_buff *skb,
+ struct iphdr *iph, struct tcphdr *tcph)
+{
+ struct sk_buff *parent = lro_desc->parent;
+ int tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
+
+ lro_add_common(lro_desc, iph, tcph, tcp_data_len);
+
+ skb_pull(skb, (skb->len - tcp_data_len));
+ parent->truesize += skb->truesize;
+
+ if (lro_desc->last_skb)
+ lro_desc->last_skb->next = skb;
+ else
+ skb_shinfo(parent)->frag_list = skb;
+
+ lro_desc->last_skb = skb;
+}
+
+static void lro_add_frags(struct net_lro_desc *lro_desc,
+ int len, int hlen, int truesize,
+ struct skb_frag_struct *skb_frags,
+ struct iphdr *iph, struct tcphdr *tcph)
+{
+ struct sk_buff *skb = lro_desc->parent;
+ int tcp_data_len = TCP_PAYLOAD_LENGTH(iph, tcph);
+
+ lro_add_common(lro_desc, iph, tcph, tcp_data_len);
+
+ skb->truesize += truesize;
+
+ skb_frags[0].page_offset += hlen;
+ skb_frags[0].size -= hlen;
+
+ while (tcp_data_len > 0) {
+ *(lro_desc->next_frag) = *skb_frags;
+ tcp_data_len -= skb_frags->size;
+ lro_desc->next_frag++;
+ skb_frags++;
+ skb_shinfo(skb)->nr_frags++;
+ }
+}
+
+static int lro_check_tcp_conn(struct net_lro_desc *lro_desc,
+ struct iphdr *iph,
+ struct tcphdr *tcph)
+{
+ if ((lro_desc->iph->saddr != iph->saddr)
+ || (lro_desc->iph->daddr != iph->daddr)
+ || (lro_desc->tcph->source != tcph->source)
+ || (lro_desc->tcph->dest != tcph->dest))
+ return -1;
+ return 0;
+}
+
+static struct net_lro_desc *lro_get_desc(struct net_lro_mgr *lro_mgr,
+ struct net_lro_desc *lro_arr,
+ struct iphdr *iph,
+ struct tcphdr *tcph)
+{
+ struct net_lro_desc *lro_desc = NULL;
+ struct net_lro_desc *tmp;
+ int max_desc = lro_mgr->max_desc;
+ int i;
+
+ for (i = 0; i < max_desc; i++) {
+ tmp = &lro_arr[i];
+ if (tmp->active)
+ if (!lro_check_tcp_conn(tmp, iph, tcph)) {
+ lro_desc = tmp;
+ goto out;
+ }
+ }
+
+ for (i = 0; i < max_desc; i++) {
+ if (!lro_arr[i].active) {
+ lro_desc = &lro_arr[i];
+ goto out;
+ }
+ }
+
+ LRO_INC_STATS(lro_mgr, no_desc);
+out:
+ return lro_desc;
+}
+
+static void lro_flush(struct net_lro_mgr *lro_mgr,
+ struct net_lro_desc *lro_desc)
+{
+ if (lro_desc->pkt_aggr_cnt > 1)
+ lro_update_tcp_ip_header(lro_desc);
+
+ if (lro_desc->vgrp) {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ vlan_hwaccel_receive_skb(lro_desc->parent,
+ lro_desc->vgrp,
+ lro_desc->vlan_tag);
+ else
+ vlan_hwaccel_rx(lro_desc->parent,
+ lro_desc->vgrp,
+ lro_desc->vlan_tag);
+
+ } else {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ netif_receive_skb(lro_desc->parent);
+ else
+ netif_rx(lro_desc->parent);
+ }
+
+ LRO_INC_STATS(lro_mgr, flushed);
+ lro_clear_desc(lro_desc);
+}
+
+static int __lro_proc_skb(struct net_lro_mgr *lro_mgr, struct sk_buff *skb,
+ struct vlan_group *vgrp, u16 vlan_tag, void *priv)
+{
+ struct net_lro_desc *lro_desc;
+ struct iphdr *iph;
+ struct tcphdr *tcph;
+ u64 flags;
+ int vlan_hdr_len = 0;
+
+ if (!lro_mgr->get_skb_header
+ || lro_mgr->get_skb_header(skb, (void *)&iph, (void *)&tcph,
+ &flags, priv))
+ goto out;
+
+ if (!(flags & LRO_IPV4) || !(flags & LRO_TCP))
+ goto out;
+
+ lro_desc = lro_get_desc(lro_mgr, lro_mgr->lro_arr, iph, tcph);
+ if (!lro_desc)
+ goto out;
+
+ if ((skb->protocol == htons(ETH_P_8021Q))
+ && !test_bit(LRO_F_EXTRACT_VLAN_ID, &lro_mgr->features))
+ vlan_hdr_len = VLAN_HLEN;
+
+ if (!lro_desc->active) { /* start new lro session */
+ if (lro_tcp_ip_check(iph, tcph, skb->len - vlan_hdr_len, NULL))
+ goto out;
+
+ lro_init_desc(lro_desc, skb, iph, tcph, vlan_tag, vgrp);
+ LRO_INC_STATS(lro_mgr, aggregated);
+ return 0;
+ }
+
+ if (lro_desc->tcp_next_seq != ntohl(tcph->seq))
+ goto out2;
+
+ if (lro_tcp_ip_check(iph, tcph, skb->len, lro_desc))
+ goto out2;
+
+ lro_add_packet(lro_desc, skb, iph, tcph);
+ LRO_INC_STATS(lro_mgr, aggregated);
+
+ if ((lro_desc->pkt_aggr_cnt >= lro_mgr->max_aggr) ||
+ lro_desc->parent->len > (0xFFFF - lro_mgr->dev->mtu))
+ lro_flush(lro_mgr, lro_desc);
+
+ return 0;
+
+out2: /* send aggregated SKBs to stack */
+ lro_flush(lro_mgr, lro_desc);
+
+out: /* Original SKB has to be posted to stack */
+ return 1;
+}
+
+
+static struct sk_buff *lro_gen_skb(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ void *mac_hdr,
+ int hlen, __wsum sum,
+ u32 ip_summed)
+{
+ struct sk_buff *skb;
+ struct skb_frag_struct *skb_frags;
+ int data_len = len;
+
+ skb = netdev_alloc_skb(lro_mgr->dev, hlen);
+ if (!skb)
+ return NULL;
+
+ skb->len = len;
+ skb->data_len = len - hlen;
+ skb->truesize += true_size;
+ skb->tail += hlen;
+
+ memcpy(skb->data, mac_hdr, hlen);
+
+ skb_frags = skb_shinfo(skb)->frags;
+ while (data_len > 0) {
+ *skb_frags = *frags;
+ data_len -= frags->size;
+ skb_frags++;
+ frags++;
+ skb_shinfo(skb)->nr_frags++;
+ }
+
+ skb_shinfo(skb)->frags[0].page_offset += hlen;
+ skb_shinfo(skb)->frags[0].size -= hlen;
+
+ skb->ip_summed = ip_summed;
+ skb->csum = sum;
+ skb->protocol = eth_type_trans(skb, lro_mgr->dev);
+ return skb;
+}
+
+static struct sk_buff *__lro_proc_segment(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ struct vlan_group *vgrp,
+ u16 vlan_tag, void *priv, __wsum sum)
+{
+ struct net_lro_desc *lro_desc;
+ struct iphdr *iph;
+ struct tcphdr *tcph;
+ struct sk_buff *skb;
+ u64 flags;
+ void *mac_hdr;
+ int mac_hdr_len;
+ int hdr_len = LRO_MAX_PG_HLEN;
+ int vlan_hdr_len = 0;
+
+ if (!lro_mgr->get_frag_header
+ || lro_mgr->get_frag_header(frags, (void *)&mac_hdr, (void *)&iph,
+ (void *)&tcph, &flags, priv)) {
+ mac_hdr = page_address(frags->page) + frags->page_offset;
+ goto out1;
+ }
+
+ if (!(flags & LRO_IPV4) || !(flags & LRO_TCP))
+ goto out1;
+
+ hdr_len = (int)((void *)(tcph) + TCP_HDR_LEN(tcph) - mac_hdr);
+ mac_hdr_len = (int)((void *)(iph) - mac_hdr);
+
+ lro_desc = lro_get_desc(lro_mgr, lro_mgr->lro_arr, iph, tcph);
+ if (!lro_desc)
+ goto out1;
+
+ if (!lro_desc->active) { /* start new lro session */
+ if (lro_tcp_ip_check(iph, tcph, len - mac_hdr_len, NULL))
+ goto out1;
+
+ skb = lro_gen_skb(lro_mgr, frags, len, true_size, mac_hdr,
+ hdr_len, 0, CHECKSUM_UNNECESSARY);
+ if (!skb)
+ goto out;
+
+ if ((skb->protocol == htons(ETH_P_8021Q))
+ && !test_bit(LRO_F_EXTRACT_VLAN_ID, &lro_mgr->features))
+ vlan_hdr_len = VLAN_HLEN;
+
+ iph = (void *)(skb->data + vlan_hdr_len);
+ tcph = (void *)((u8 *)skb->data + IP_HDR_LEN(iph));
+
+ lro_init_desc(lro_desc, skb, iph, tcph, 0, NULL);
+ LRO_INC_STATS(lro_mgr, aggregated);
+ return 0;
+ }
+
+ if (lro_desc->tcp_next_seq != ntohl(tcph->seq))
+ goto out2;
+
+ if (lro_tcp_ip_check(iph, tcph, len - mac_hdr_len, lro_desc))
+ goto out2;
+
+ lro_add_frags(lro_desc, len, hdr_len, true_size, frags, iph, tcph);
+ LRO_INC_STATS(lro_mgr, aggregated);
+
+ if ((skb_shinfo(lro_desc->parent)->nr_frags >= lro_mgr->max_aggr) ||
+ lro_desc->parent->len > (0xFFFF - lro_mgr->dev->mtu))
+ lro_flush(lro_mgr, lro_desc);
+
+ return NULL;
+
+out2: /* send aggregated packets to the stack */
+ lro_flush(lro_mgr, lro_desc);
+
+out1: /* Original packet has to be posted to the stack */
+ skb = lro_gen_skb(lro_mgr, frags, len, true_size, mac_hdr,
+ hdr_len, sum, lro_mgr->ip_summed);
+out:
+ return skb;
+}
+
+void lro_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ void *priv)
+{
+ if (__lro_proc_skb(lro_mgr, skb, NULL, 0, priv)) {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ netif_receive_skb(skb);
+ else
+ netif_rx(skb);
+ }
+}
+EXPORT_SYMBOL(lro_receive_skb);
+
+void lro_vlan_hwaccel_receive_skb(struct net_lro_mgr *lro_mgr,
+ struct sk_buff *skb,
+ struct vlan_group *vgrp,
+ u16 vlan_tag,
+ void *priv)
+{
+ if (__lro_proc_skb(lro_mgr, skb, vgrp, vlan_tag, priv)) {
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ vlan_hwaccel_receive_skb(skb, vgrp, vlan_tag);
+ else
+ vlan_hwaccel_rx(skb, vgrp, vlan_tag);
+ }
+}
+EXPORT_SYMBOL(lro_vlan_hwaccel_receive_skb);
+
+void lro_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size, void *priv, __wsum sum)
+{
+ struct sk_buff *skb;
+
+ skb = __lro_proc_segment(lro_mgr, frags, len, true_size, NULL, 0,
+ priv, sum);
+ if (!skb)
+ return;
+
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ netif_receive_skb(skb);
+ else
+ netif_rx(skb);
+}
+EXPORT_SYMBOL(lro_receive_frags);
+
+void lro_vlan_hwaccel_receive_frags(struct net_lro_mgr *lro_mgr,
+ struct skb_frag_struct *frags,
+ int len, int true_size,
+ struct vlan_group *vgrp,
+ u16 vlan_tag, void *priv, __wsum sum)
+{
+ struct sk_buff *skb;
+
+ skb = __lro_proc_segment(lro_mgr, frags, len, true_size, vgrp,
+ vlan_tag, priv, sum);
+ if (!skb)
+ return;
+
+ if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ vlan_hwaccel_receive_skb(skb, vgrp, vlan_tag);
+ else
+ vlan_hwaccel_rx(skb, vgrp, vlan_tag);
+}
+EXPORT_SYMBOL(lro_vlan_hwaccel_receive_frags);
+
+void lro_flush_all(struct net_lro_mgr *lro_mgr)
+{
+ int i;
+ struct net_lro_desc *lro_desc = lro_mgr->lro_arr;
+
+ for (i = 0; i < lro_mgr->max_desc; i++) {
+ if (lro_desc[i].active)
+ lro_flush(lro_mgr, &lro_desc[i]);
+ }
+}
+EXPORT_SYMBOL(lro_flush_all);
+
+void lro_flush_pkt(struct net_lro_mgr *lro_mgr,
+ struct iphdr *iph, struct tcphdr *tcph)
+{
+ struct net_lro_desc *lro_desc;
+
+ lro_desc = lro_get_desc(lro_mgr, lro_mgr->lro_arr, iph, tcph);
+ if (lro_desc->active)
+ lro_flush(lro_mgr, lro_desc);
+}
+EXPORT_SYMBOL(lro_flush_pkt);
--
1.5.2
^ permalink raw reply related
* [PATCH 2/4][RFC] lro: Kconfig and Makefile
From: Jan-Bernd Themann @ 2007-07-30 15:24 UTC (permalink / raw)
To: netdev
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, Stefan Roscher, David Miller,
Andrew Gallatin, Jeff Garzik
Kconfig and Makefile for LRO
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
---
net/ipv4/Kconfig | 8 ++++++++
net/ipv4/Makefile | 1 +
2 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index fb79097..d894f61 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -394,6 +394,14 @@ config INET_XFRM_MODE_BEET
If unsure, say Y.
+config INET_LRO
+ tristate "Large Receive Offload (ipv4/tcp)"
+
+ ---help---
+ Support for Large Receive Offload (ipv4/tcp).
+
+ If unsure, say Y.
+
config INET_DIAG
tristate "INET: socket monitoring interface"
default y
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index fbf1674..a02c36d 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_INET_ESP) += esp4.o
obj-$(CONFIG_INET_IPCOMP) += ipcomp.o
obj-$(CONFIG_INET_XFRM_TUNNEL) += xfrm4_tunnel.o
obj-$(CONFIG_INET_XFRM_MODE_BEET) += xfrm4_mode_beet.o
+obj-$(CONFIG_INET_LRO) += inet_lro.o
obj-$(CONFIG_INET_TUNNEL) += tunnel4.o
obj-$(CONFIG_INET_XFRM_MODE_TRANSPORT) += xfrm4_mode_transport.o
obj-$(CONFIG_INET_XFRM_MODE_TUNNEL) += xfrm4_mode_tunnel.o
--
1.5.2
^ permalink raw reply related
* [PATCH 0/4][RFC] lro: Generic Large Receive Offload for TCP traffic
From: Jan-Bernd Themann @ 2007-07-30 15:24 UTC (permalink / raw)
To: netdev
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, Stefan Roscher, David Miller,
Andrew Gallatin, Jeff Garzik
Hi,
this patch set contains the latest generic LRO code, a Kconfig / Makefile
and an eHEA patch demonstrating how the "aggregate SKB" interface has to
to be used.
Drew, could you provide a patch for the myri10ge driver to show how the
"receive in page" interface works?
Please check the Kconfig / Makefile patch. Is that the right place for
the LRO entries?
There is still one open question for the "receive in page" mode:
How many data (length) has to be copied to skb->data for packets that
do not work for LRO (other protocols?). Currently I choose 64 as default.
Is that ok?
Thanks,
Jan-Bernd
[PATCH 1/4][RFC] lro: Generic Large Receive Offload for TCP traffic
[PATCH 2/4][RFC] lro: Kconfig and Makefile
[PATCH 3/4][RFC] ehea: LRO support
[PATCH 4/4][RFC] ehea: Kconfig
Changes to http://www.spinics.net/lists/netdev/msg36912.html
1) A new field called "features" has been added to the net_lro_mgr struct.
It is set by the driver to indicate:
- LRO_F_NAPI: Use NAPI / netif_rx to pass packets to stack
- LRO_F_EXTRACT_VLAN_ID: Set by driver if HW extracts VLAN IDs for VLAN
packets but does not modify ETH protocol (ETH_P_8021Q)
2) Padded frames are not aggregated for now. Bug fixed
3) Correct header length now used. No minimal header length for aggregated
packets used anymore.
4) Statistic counters were introduced. They are stored in a new struct in
the net_lro_mgr. This has the advantage that no locking is required in
cases where the driver uses multiple lro_mgrs for different receive queues.
Thus we get the following statistics per lro_mgr / eth device:
- Number of aggregated packets
- Number of flushed packets
- Number of times we run out of lro_desc.
The ratio of "aggregated packets" and "flushed packets" give you an
idea how well LRO is working.
^ permalink raw reply
* [PATCH 4/4][RFC] ehea: Kconfig
From: Jan-Bernd Themann @ 2007-07-30 15:24 UTC (permalink / raw)
To: netdev
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, Stefan Roscher, David Miller,
Andrew Gallatin, Jeff Garzik
Kconfig changes for LRO
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
---
drivers/net/Kconfig | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index f8a602c..fec4004 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2399,6 +2399,7 @@ config CHELSIO_T3
config EHEA
tristate "eHEA Ethernet support"
depends on IBMEBUS
+ select INET_LRO
---help---
This driver supports the IBM pSeries eHEA ethernet adapter.
--
1.5.2
^ permalink raw reply related
* [PATCH 3/4][RFC] ehea: LRO support
From: Jan-Bernd Themann @ 2007-07-30 15:24 UTC (permalink / raw)
To: netdev
Cc: Christoph Raisch, Jan-Bernd Themann, linux-kernel, linux-ppc,
Marcus Eder, Thomas Klein, Stefan Roscher, Jeff Garzik,
David Miller, Andrew Gallatin
Added LRO support using the "SKB aggregate" interface
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
---
drivers/net/ehea/ehea.h | 9 ++++-
drivers/net/ehea/ehea_ethtool.c | 15 +++++++
drivers/net/ehea/ehea_main.c | 82 +++++++++++++++++++++++++++++++++++---
3 files changed, 98 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h
index d67f97b..70e33fe 100644
--- a/drivers/net/ehea/ehea.h
+++ b/drivers/net/ehea/ehea.h
@@ -33,13 +33,14 @@
#include <linux/ethtool.h>
#include <linux/vmalloc.h>
#include <linux/if_vlan.h>
+#include <linux/inet_lro.h>
#include <asm/ibmebus.h>
#include <asm/abs_addr.h>
#include <asm/io.h>
#define DRV_NAME "ehea"
-#define DRV_VERSION "EHEA_0073"
+#define DRV_VERSION "EHEA_0074"
/* eHEA capability flags */
#define DLPAR_PORT_ADD_REM 1
@@ -58,6 +59,7 @@
#define EHEA_SMALL_QUEUES
#define EHEA_NUM_TX_QP 1
+#define EHEA_LRO_MAX_AGGR 64
#ifdef EHEA_SMALL_QUEUES
#define EHEA_MAX_CQE_COUNT 1023
@@ -84,6 +86,8 @@
#define EHEA_RQ2_PKT_SIZE 1522
#define EHEA_L_PKT_SIZE 256 /* low latency */
+#define MAX_LRO_DESCRIPTORS 8
+
/* Send completion signaling */
/* Protection Domain Identifier */
@@ -376,6 +380,8 @@ struct ehea_port_res {
u64 tx_packets;
u64 rx_packets;
u32 poll_counter;
+ struct net_lro_mgr lro_mgr;
+ struct net_lro_desc lro_desc[MAX_LRO_DESCRIPTORS];
};
@@ -427,6 +433,7 @@ struct ehea_port {
u32 msg_enable;
u32 sig_comp_iv;
u32 state;
+ u32 lro_max_aggr;
u8 full_duplex;
u8 autoneg;
u8 num_def_qps;
diff --git a/drivers/net/ehea/ehea_ethtool.c b/drivers/net/ehea/ehea_ethtool.c
index decec8c..29ef7a9 100644
--- a/drivers/net/ehea/ehea_ethtool.c
+++ b/drivers/net/ehea/ehea_ethtool.c
@@ -183,6 +183,9 @@ static char ehea_ethtool_stats_keys[][ETH_GSTRING_LEN] = {
{"PR5 free_swqes"},
{"PR6 free_swqes"},
{"PR7 free_swqes"},
+ {"LRO aggregated"},
+ {"LRO flushed"},
+ {"LRO no_desc"},
};
static void ehea_get_strings(struct net_device *dev, u32 stringset, u8 *data)
@@ -239,6 +242,18 @@ static void ehea_get_ethtool_stats(struct net_device *dev,
for (k = 0; k < 8; k++)
data[i++] = atomic_read(&port->port_res[k].swqe_avail);
+ for (k = 0, tmp = 0; k < EHEA_MAX_PORT_RES; k++)
+ tmp |= port->port_res[k].lro_mgr.stats.aggregated;
+ data[i++] = tmp;
+
+ for (k = 0, tmp = 0; k < EHEA_MAX_PORT_RES; k++)
+ tmp |= port->port_res[k].lro_mgr.stats.flushed;
+ data[i++] = tmp;
+
+ for (k = 0, tmp = 0; k < EHEA_MAX_PORT_RES; k++)
+ tmp |= port->port_res[k].lro_mgr.stats.no_desc;
+ data[i++] = tmp;
+
}
const struct ethtool_ops ehea_ethtool_ops = {
diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c
index 9756211..41bc075 100644
--- a/drivers/net/ehea/ehea_main.c
+++ b/drivers/net/ehea/ehea_main.c
@@ -52,6 +52,8 @@ static int rq2_entries = EHEA_DEF_ENTRIES_RQ2;
static int rq3_entries = EHEA_DEF_ENTRIES_RQ3;
static int sq_entries = EHEA_DEF_ENTRIES_SQ;
static int use_mcs = 0;
+static int use_lro = 0;
+static int lro_max_aggr = EHEA_LRO_MAX_AGGR;
static int num_tx_qps = EHEA_NUM_TX_QP;
module_param(msg_level, int, 0);
@@ -60,6 +62,8 @@ module_param(rq2_entries, int, 0);
module_param(rq3_entries, int, 0);
module_param(sq_entries, int, 0);
module_param(use_mcs, int, 0);
+module_param(use_lro, int, 0);
+module_param(lro_max_aggr, int, 0);
module_param(num_tx_qps, int, 0);
MODULE_PARM_DESC(num_tx_qps, "Number of TX-QPS");
@@ -77,6 +81,10 @@ MODULE_PARM_DESC(sq_entries, " Number of entries for the Send Queue "
"[2^x - 1], x = [6..14]. Default = "
__MODULE_STRING(EHEA_DEF_ENTRIES_SQ) ")");
MODULE_PARM_DESC(use_mcs, " 0:NAPI, 1:Multiple receive queues, Default = 1 ");
+MODULE_PARM_DESC(lro_max_aggr, " LRO: Max packets to be aggregated. Default = "
+ __MODULE_STRING(EHEA_LRO_MAX_AGGR));
+MODULE_PARM_DESC(use_lro, " Large Receive Offload, 1: enable, 0: disable, "
+ "Default = 0");
static int port_name_cnt = 0;
static LIST_HEAD(adapter_list);
@@ -389,6 +397,60 @@ static int ehea_treat_poll_error(struct ehea_port_res *pr, int rq,
return 0;
}
+static int get_skb_hdr(struct sk_buff *skb, void **iphdr,
+ void **tcph, u64 *hdr_flags, void *priv)
+{
+ struct ehea_cqe *cqe = priv;
+ unsigned int ip_len;
+ struct iphdr *iph;
+
+ /* non tcp/udp packets */
+ if (!cqe->header_length)
+ return -1;
+
+ /* non tcp packet */
+ skb_reset_network_header(skb);
+ iph = ip_hdr(skb);
+ if (iph->protocol != IPPROTO_TCP)
+ return -1;
+
+ ip_len = ip_hdrlen(skb);
+ skb_set_transport_header(skb, ip_len);
+ *tcph = tcp_hdr(skb);
+
+ /* check if ip header and tcp header are complete */
+ if (iph->tot_len < ip_len + tcp_hdrlen(skb))
+ return -1;
+
+ *hdr_flags = LRO_IPV4 | LRO_TCP;
+ *iphdr = iph;
+
+ return 0;
+}
+
+static void ehea_proc_skb(struct ehea_port_res *pr, struct ehea_cqe *cqe,
+ struct sk_buff *skb)
+{
+ int vlan_extracted = (cqe->status & EHEA_CQE_VLAN_TAG_XTRACT)
+ && pr->port->vgrp;
+
+ if (use_lro) {
+ if (vlan_extracted)
+ lro_vlan_hwaccel_receive_skb(&pr->lro_mgr, skb,
+ pr->port->vgrp,
+ cqe->vlan_tag,
+ cqe);
+ else
+ lro_receive_skb(&pr->lro_mgr, skb, cqe);
+ } else {
+ if (vlan_extracted)
+ vlan_hwaccel_receive_skb(skb, pr->port->vgrp,
+ cqe->vlan_tag);
+ else
+ netif_receive_skb(skb);
+ }
+}
+
static struct ehea_cqe *ehea_proc_rwqes(struct net_device *dev,
struct ehea_port_res *pr,
int *budget)
@@ -460,13 +522,7 @@ static struct ehea_cqe *ehea_proc_rwqes(struct net_device *dev,
processed_rq3++;
}
- if ((cqe->status & EHEA_CQE_VLAN_TAG_XTRACT)
- && port->vgrp)
- vlan_hwaccel_receive_skb(skb, port->vgrp,
- cqe->vlan_tag);
- else
- netif_receive_skb(skb);
-
+ ehea_proc_skb(pr, cqe, skb);
dev->last_rx = jiffies;
} else {
pr->p_stats.poll_receive_errors++;
@@ -478,6 +534,8 @@ static struct ehea_cqe *ehea_proc_rwqes(struct net_device *dev,
}
cqe = ehea_poll_rq1(qp, &wqe_index);
}
+ if (use_lro)
+ lro_flush_all(&pr->lro_mgr);
pr->rx_packets += processed;
*budget -= processed;
@@ -1233,6 +1291,13 @@ static int ehea_init_port_res(struct ehea_port *port, struct ehea_port_res *pr,
set_bit(__LINK_STATE_START, &pr->d_netdev->state);
strcpy(pr->d_netdev->name, port->netdev->name);
+ pr->lro_mgr.max_aggr = pr->port->lro_max_aggr;
+ pr->lro_mgr.max_desc = MAX_LRO_DESCRIPTORS;
+ pr->lro_mgr.lro_arr = pr->lro_desc;
+ pr->lro_mgr.get_skb_header = get_skb_hdr;
+ pr->lro_mgr.features = LRO_F_NAPI | LRO_F_EXTRACT_VLAN_ID;
+ pr->lro_mgr.dev = port->netdev;
+
ret = 0;
goto out;
@@ -1712,6 +1777,7 @@ static int ehea_change_mtu(struct net_device *dev, int new_mtu)
if ((new_mtu < 68) || (new_mtu > EHEA_MAX_PACKET_SIZE))
return -EINVAL;
dev->mtu = new_mtu;
+
return 0;
}
@@ -2669,6 +2735,8 @@ struct ehea_port *ehea_setup_single_port(struct ehea_adapter *adapter,
goto out_dereg_bc;
}
+ port->lro_max_aggr = lro_max_aggr;
+
ret = ehea_get_jumboframe_status(port, &jumbo);
if (ret)
ehea_error("failed determining jumbo frame status for %s",
--
1.5.2
^ permalink raw reply related
* Re: source interface ping bug ?
From: Ben Greear @ 2007-07-30 16:06 UTC (permalink / raw)
To: nano bug; +Cc: netdev
In-Reply-To: <d39c36500707300146h72c72d69rce2f68595bdf6699@mail.gmail.com>
nano bug wrote:
> Can someone have a look a this and tell if it's kernel related or if I
> posted this in the wrong place ? Thanks.
>
Last I checked, ping did not do an SO_BINDTODEVICE even if you did -i ethX.
I think it just looked up the IP for that port and treated it as -i a.b.c.d.
That said, I'm not sure why the behaviour changes for you between kernel
releases.
Maybe an 'strace' of your ping command on the different kernels would help
figure out what the problem is?
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* [PATCH] [NET]: fix multicast list when cloning sockets
From: Flavio Leitner @ 2007-07-30 16:04 UTC (permalink / raw)
To: David L Stevens; +Cc: davem, acme, netdev
The sock_copy() function uses memcpy() to clone the socket
including the struct ip_mc_socklist *mc_list pointer.
The ip_mc_drop_socket() function is called when socket is closed
to free these objects leaving the other sockets cloned from the
same master socket with invalid pointers.
This patch sets mc_list of cloned socket to NULL.
Signed-off-by: Flavio Leitner <fleitner@redhat.com>
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index fbe7714..8ee0f54 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -506,6 +506,8 @@ struct sock *inet_csk_clone(struct sock *sk, const struct request_sock *req,
newicsk->icsk_backoff = 0;
newicsk->icsk_probes_out = 0;
+ inet_sk(inet)->mc_list = NULL;
+
/* Deinitialize accept_queue to trap illegal accesses. */
memset(&newicsk->icsk_accept_queue, 0, sizeof(newicsk->icsk_accept_queue));
--
1.5.2.4
--
Flavio
^ permalink raw reply related
* Re: [PATCH 0/4][RFC] lro: Generic Large Receive Offload for TCP traffic
From: Jeff Garzik @ 2007-07-30 16:17 UTC (permalink / raw)
To: Jan-Bernd Themann, David Miller
Cc: Thomas Klein, Jan-Bernd Themann, netdev, linux-kernel, linux-ppc,
Christoph Raisch, Marcus Eder, Andrew Gallatin, Stefan Roscher
In-Reply-To: <200707301724.33865.ossthema@de.ibm.com>
Seems pretty good to me, save for one minor detail: patches #1/#2
should be combined together for greater git-bisect happiness. Ditto for
patches #3/#4. Largely harmless in this case, but keeps the git history
pollution to a minimum.
Caveat reviewer: I'm not an expert of net/ipv4/* code, so I reviewed
largely from the driver API perspective.
David, thoughts on merging? I'm not We could stick this into your tree
or mine. Whether yours or mine, I would like to keep the driver and
net-core patches together in the same git tree.
Jeff
^ permalink raw reply
* Re: source interface ping bug ?
From: Patrick McHardy @ 2007-07-30 16:19 UTC (permalink / raw)
To: nano bug; +Cc: netdev
In-Reply-To: <d39c36500707270130j3eac7f44qb3a4c87496a1296b@mail.gmail.com>
nano bug wrote:
> [...]
> using source interface :
>
> root@darkstar:~/iputils# ./ping -I eth2 87.248.113.14
> PING 87.248.113.14 (87.248.113.14) from 86.106.19.75 eth2: 56(84) bytes of data.
>>From 86.106.19.75 icmp_seq=1 Destination Host Unreachable
> root@darkstar:~# tcpdump -i eth2 -vvv -n host 87.248.113.14 and host
> 86.106.19.75
> tcpdump: listening on eth2, link-type EN10MB (Ethernet), capture size 96 bytes
> 01:19:24.292911 arp who-has 87.248.113.14 tell 86.106.19.75
Are you using (or running) NAT locally? What do your routing tables look
like?
^ permalink raw reply
* Re: [PATCH 2/4][RFC] lro: Kconfig and Makefile
From: Stephen Hemminger @ 2007-07-30 16:25 UTC (permalink / raw)
To: Jan-Bernd Themann
Cc: netdev, Christoph Raisch, Jan-Bernd Themann, linux-kernel,
linux-ppc, Marcus Eder, Thomas Klein, Stefan Roscher,
David Miller, Andrew Gallatin, Jeff Garzik
In-Reply-To: <200707301724.45781.ossthema@de.ibm.com>
On Mon, 30 Jul 2007 17:24:45 +0200
Jan-Bernd Themann <ossthema@de.ibm.com> wrote:
> Kconfig and Makefile for LRO
>
> Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
>
> ---
> net/ipv4/Kconfig | 8 ++++++++
> net/ipv4/Makefile | 1 +
> 2 files changed, 9 insertions(+), 0 deletions(-)
>
> diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
> index fb79097..d894f61 100644
> --- a/net/ipv4/Kconfig
> +++ b/net/ipv4/Kconfig
> @@ -394,6 +394,14 @@ config INET_XFRM_MODE_BEET
>
> If unsure, say Y.
>
> +config INET_LRO
> + tristate "Large Receive Offload (ipv4/tcp)"
> +
> + ---help---
> + Support for Large Receive Offload (ipv4/tcp).
> +
> + If unsure, say Y.
> +
Why make this a user selectable option at all? Unless you want
to deal with out of tree drivers (not my problem), it should be hidden
to avoid having to explain an support it.
^ 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