* [patch net-next 04/15] rtnetlink: remove usage of dev->master
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
net/core/rtnetlink.c | 69 ++++++++++++++++++++++++++++------------------------
1 file changed, 37 insertions(+), 32 deletions(-)
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 2ef7a56..ae612f4 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -880,6 +880,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
const struct rtnl_link_stats64 *stats;
struct nlattr *attr, *af_spec;
struct rtnl_af_ops *af_ops;
+ struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
ASSERT_RTNL();
nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags);
@@ -908,8 +909,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
#endif
(dev->ifindex != dev->iflink &&
nla_put_u32(skb, IFLA_LINK, dev->iflink)) ||
- (dev->master &&
- nla_put_u32(skb, IFLA_MASTER, dev->master->ifindex)) ||
+ (upper_dev &&
+ nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex)) ||
nla_put_u8(skb, IFLA_CARRIER, netif_carrier_ok(dev)) ||
(dev->qdisc &&
nla_put_string(skb, IFLA_QDISC, dev->qdisc->ops->id)) ||
@@ -1273,16 +1274,16 @@ static int do_setvfinfo(struct net_device *dev, struct nlattr *attr)
static int do_set_master(struct net_device *dev, int ifindex)
{
- struct net_device *master_dev;
+ struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
const struct net_device_ops *ops;
int err;
- if (dev->master) {
- if (dev->master->ifindex == ifindex)
+ if (upper_dev) {
+ if (upper_dev->ifindex == ifindex)
return 0;
- ops = dev->master->netdev_ops;
+ ops = upper_dev->netdev_ops;
if (ops->ndo_del_slave) {
- err = ops->ndo_del_slave(dev->master, dev);
+ err = ops->ndo_del_slave(upper_dev, dev);
if (err)
return err;
} else {
@@ -1291,12 +1292,12 @@ static int do_set_master(struct net_device *dev, int ifindex)
}
if (ifindex) {
- master_dev = __dev_get_by_index(dev_net(dev), ifindex);
- if (!master_dev)
+ upper_dev = __dev_get_by_index(dev_net(dev), ifindex);
+ if (!upper_dev)
return -EINVAL;
- ops = master_dev->netdev_ops;
+ ops = upper_dev->netdev_ops;
if (ops->ndo_add_slave) {
- err = ops->ndo_add_slave(master_dev, dev);
+ err = ops->ndo_add_slave(upper_dev, dev);
if (err)
return err;
} else {
@@ -2064,7 +2065,6 @@ errout:
static int rtnl_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
struct net *net = sock_net(skb->sk);
- struct net_device *master = NULL;
struct ndmsg *ndm;
struct nlattr *tb[NDA_MAX+1];
struct net_device *dev;
@@ -2106,10 +2106,10 @@ static int rtnl_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
/* Support fdb on master device the net/bridge default case */
if ((!ndm->ndm_flags || ndm->ndm_flags & NTF_MASTER) &&
(dev->priv_flags & IFF_BRIDGE_PORT)) {
- master = dev->master;
- err = master->netdev_ops->ndo_fdb_add(ndm, tb,
- dev, addr,
- nlh->nlmsg_flags);
+ struct net_device *br_dev = netdev_master_upper_dev_get(dev);
+ const struct net_device_ops *ops = br_dev->netdev_ops;
+
+ err = ops->ndo_fdb_add(ndm, tb, dev, addr, nlh->nlmsg_flags);
if (err)
goto out;
else
@@ -2170,10 +2170,11 @@ static int rtnl_fdb_del(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
/* Support fdb on master device the net/bridge default case */
if ((!ndm->ndm_flags || ndm->ndm_flags & NTF_MASTER) &&
(dev->priv_flags & IFF_BRIDGE_PORT)) {
- struct net_device *master = dev->master;
+ struct net_device *br_dev = netdev_master_upper_dev_get(dev);
+ const struct net_device_ops *ops = br_dev->netdev_ops;
- if (master->netdev_ops->ndo_fdb_del)
- err = master->netdev_ops->ndo_fdb_del(ndm, dev, addr);
+ if (ops->ndo_fdb_del)
+ err = ops->ndo_fdb_del(ndm, dev, addr);
if (err)
goto out;
@@ -2257,9 +2258,11 @@ static int rtnl_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
rcu_read_lock();
for_each_netdev_rcu(net, dev) {
if (dev->priv_flags & IFF_BRIDGE_PORT) {
- struct net_device *master = dev->master;
- const struct net_device_ops *ops = master->netdev_ops;
+ struct net_device *br_dev;
+ const struct net_device_ops *ops;
+ br_dev = netdev_master_upper_dev_get(dev);
+ ops = br_dev->netdev_ops;
if (ops->ndo_fdb_dump)
idx = ops->ndo_fdb_dump(skb, cb, dev, idx);
}
@@ -2280,6 +2283,7 @@ int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
struct ifinfomsg *ifm;
struct nlattr *br_afspec;
u8 operstate = netif_running(dev) ? dev->operstate : IF_OPER_DOWN;
+ struct net_device *br_dev = netdev_master_upper_dev_get(dev);
nlh = nlmsg_put(skb, pid, seq, RTM_NEWLINK, sizeof(*ifm), NLM_F_MULTI);
if (nlh == NULL)
@@ -2297,8 +2301,8 @@ int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
if (nla_put_string(skb, IFLA_IFNAME, dev->name) ||
nla_put_u32(skb, IFLA_MTU, dev->mtu) ||
nla_put_u8(skb, IFLA_OPERSTATE, operstate) ||
- (dev->master &&
- nla_put_u32(skb, IFLA_MASTER, dev->master->ifindex)) ||
+ (br_dev &&
+ nla_put_u32(skb, IFLA_MASTER, br_dev->ifindex)) ||
(dev->addr_len &&
nla_put(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr)) ||
(dev->ifindex != dev->iflink &&
@@ -2334,11 +2338,11 @@ static int rtnl_bridge_getlink(struct sk_buff *skb, struct netlink_callback *cb)
rcu_read_lock();
for_each_netdev_rcu(net, dev) {
const struct net_device_ops *ops = dev->netdev_ops;
- struct net_device *master = dev->master;
+ struct net_device *br_dev = netdev_master_upper_dev_get(dev);
- if (master && master->netdev_ops->ndo_bridge_getlink) {
+ if (br_dev && br_dev->netdev_ops->ndo_bridge_getlink) {
if (idx >= cb->args[0] &&
- master->netdev_ops->ndo_bridge_getlink(
+ br_dev->netdev_ops->ndo_bridge_getlink(
skb, portid, seq, dev) < 0)
break;
idx++;
@@ -2375,7 +2379,7 @@ static inline size_t bridge_nlmsg_size(void)
static int rtnl_bridge_notify(struct net_device *dev, u16 flags)
{
struct net *net = dev_net(dev);
- struct net_device *master = dev->master;
+ struct net_device *br_dev = netdev_master_upper_dev_get(dev);
struct sk_buff *skb;
int err = -EOPNOTSUPP;
@@ -2386,8 +2390,8 @@ static int rtnl_bridge_notify(struct net_device *dev, u16 flags)
}
if ((!flags || (flags & BRIDGE_FLAGS_MASTER)) &&
- master && master->netdev_ops->ndo_bridge_getlink) {
- err = master->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev);
+ br_dev && br_dev->netdev_ops->ndo_bridge_getlink) {
+ err = br_dev->netdev_ops->ndo_bridge_getlink(skb, 0, 0, dev);
if (err < 0)
goto errout;
}
@@ -2446,13 +2450,14 @@ static int rtnl_bridge_setlink(struct sk_buff *skb, struct nlmsghdr *nlh,
oflags = flags;
if (!flags || (flags & BRIDGE_FLAGS_MASTER)) {
- if (!dev->master ||
- !dev->master->netdev_ops->ndo_bridge_setlink) {
+ struct net_device *br_dev = netdev_master_upper_dev_get(dev);
+
+ if (!br_dev || !br_dev->netdev_ops->ndo_bridge_setlink) {
err = -EOPNOTSUPP;
goto out;
}
- err = dev->master->netdev_ops->ndo_bridge_setlink(dev, nlh);
+ err = br_dev->netdev_ops->ndo_bridge_setlink(dev, nlh);
if (err)
goto out;
--
1.8.0
^ permalink raw reply related
* [patch net-next 05/15] team: remove usage of netdev_set_master()
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
drivers/net/team/team.c | 13 +++++++------
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 918a901..3df5cc1 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -1055,10 +1055,11 @@ static int team_port_add(struct team *team, struct net_device *port_dev)
}
}
- err = netdev_set_master(port_dev, dev);
+ err = netdev_master_upper_dev_link(port_dev, dev);
if (err) {
- netdev_err(dev, "Device %s failed to set master\n", portname);
- goto err_set_master;
+ netdev_err(dev, "Device %s failed to set upper link\n",
+ portname);
+ goto err_set_upper_link;
}
err = netdev_rx_handler_register(port_dev, team_handle_frame,
@@ -1091,9 +1092,9 @@ err_option_port_add:
netdev_rx_handler_unregister(port_dev);
err_handler_register:
- netdev_set_master(port_dev, NULL);
+ netdev_upper_dev_unlink(port_dev, dev);
-err_set_master:
+err_set_upper_link:
team_port_disable_netpoll(port);
err_enable_netpoll:
@@ -1137,7 +1138,7 @@ static int team_port_del(struct team *team, struct net_device *port_dev)
team_port_disable(team, port);
list_del_rcu(&port->list);
netdev_rx_handler_unregister(port_dev);
- netdev_set_master(port_dev, NULL);
+ netdev_upper_dev_unlink(port_dev, dev);
team_port_disable_netpoll(port);
vlan_vids_del_by_dev(port_dev, dev);
dev_close(port_dev);
--
1.8.0
^ permalink raw reply related
* [patch net-next 06/15] bridge: remove usage of netdev_set_master()
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
net/bridge/br_if.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 1edd71d..2148d47 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -148,7 +148,7 @@ static void del_nbp(struct net_bridge_port *p)
netdev_rx_handler_unregister(dev);
synchronize_net();
- netdev_set_master(dev, NULL);
+ netdev_upper_dev_unlink(dev, br->dev);
br_multicast_del_port(p);
@@ -364,7 +364,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
if (br_netpoll_info(br) && ((err = br_netpoll_enable(p, GFP_KERNEL))))
goto err3;
- err = netdev_set_master(dev, br->dev);
+ err = netdev_master_upper_dev_link(dev, br->dev);
if (err)
goto err4;
@@ -403,7 +403,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
return 0;
err5:
- netdev_set_master(dev, NULL);
+ netdev_upper_dev_unlink(dev, br->dev);
err4:
br_netpoll_disable(p);
err3:
--
1.8.0
^ permalink raw reply related
* [patch net-next 07/15] netpoll: remove usage of dev->master
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-git-send-email-jiri@resnulli.us>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
---
net/core/netpoll.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 3151acf..d2bda8e 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -210,9 +210,12 @@ static void netpoll_poll_dev(struct net_device *dev)
if (dev->flags & IFF_SLAVE) {
if (ni) {
- struct net_device *bond_dev = dev->master;
+ struct net_device *bond_dev;
struct sk_buff *skb;
- struct netpoll_info *bond_ni = rcu_dereference_bh(bond_dev->npinfo);
+ struct netpoll_info *bond_ni;
+
+ bond_dev = netdev_master_upper_dev_get_rcu(dev);
+ bond_ni = rcu_dereference_bh(bond_dev->npinfo);
while ((skb = skb_dequeue(&ni->arp_tx))) {
skb->dev = bond_dev;
skb_queue_tail(&bond_ni->arp_tx, skb);
@@ -815,7 +818,7 @@ int netpoll_setup(struct netpoll *np)
return -ENODEV;
}
- if (ndev->master) {
+ if (netdev_master_upper_dev_get(ndev)) {
np_err(np, "%s is a slave device, aborting\n", np->dev_name);
err = -EBUSY;
goto put;
--
1.8.0
^ permalink raw reply related
* [patch net-next 08/15] cxgb3: remove usage of dev->master
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 942dace..3f1f501 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_master_upper_dev_get_rcu(dev)))
+ dev = upper_dev;
}
+ rcu_read_unlock();
return dev;
}
}
--
1.8.0
^ permalink raw reply related
* [patch net-next 09/15] qlcnic: guard __vlan_find_dev_deep() by rcu_read_lock
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 d833f59..95b628f 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -2867,12 +2867,14 @@ 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,
--
1.8.0
^ permalink raw reply related
* [patch net-next 10/15] qeth: ensure that __vlan_find_dev_deep() is called with rcu_read_lock
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 6e5eef0..0749efe 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1640,6 +1640,7 @@ 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;
@@ -1652,19 +1653,14 @@ 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);
}
}
@@ -1673,14 +1669,14 @@ 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
@@ -1705,6 +1701,7 @@ 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;
@@ -1741,10 +1738,12 @@ 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 */
@@ -1813,8 +1812,10 @@ 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)
--
1.8.0
^ permalink raw reply related
* [patch net-next 11/15] vlan: remove usage of dev->master in __vlan_find_dev_deep()
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 65e06ab..380440b 100644
--- a/net/8021q/vlan_core.c
+++ b/net/8021q/vlan_core.c
@@ -60,21 +60,25 @@ bool vlan_do_receive(struct sk_buff **skbp)
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 master 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_master_upper_dev_get_rcu(dev);
+ if (upper_dev)
+ return __vlan_find_dev_deep(upper_dev, vlan_id);
}
return NULL;
--
1.8.0
^ permalink raw reply related
* [patch net-next 12/15] nes: remove usage of dev->master
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 748db2d..d129c36 100644
--- a/drivers/infiniband/hw/nes/nes.c
+++ b/drivers/infiniband/hw/nes/nes.c
@@ -135,6 +135,7 @@ 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;
@@ -145,8 +146,9 @@ 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_master_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"
@@ -179,9 +181,9 @@ 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 22ea67e..ca6fbae 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -1340,7 +1340,7 @@ 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;
--
1.8.0
^ permalink raw reply related
* [patch net-next 13/15] bonding: remove usage of dev->master
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 7c9d136..f5e0527 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -507,7 +507,7 @@ 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;
}
@@ -517,7 +517,7 @@ 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;
}
}
@@ -1043,7 +1043,7 @@ 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 b7d45f3..c495298 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -746,11 +746,9 @@ 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();
}
/*
@@ -760,9 +758,10 @@ 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;
@@ -774,18 +773,14 @@ 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_master_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);
}
@@ -794,13 +789,16 @@ 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();
}
/*
@@ -1493,6 +1491,27 @@ static rx_handler_result_t bond_handle_frame(struct sk_buff **pskb)
return ret;
}
+static int bond_master_upper_dev_link(struct net_device *bond_dev,
+ struct net_device *slave_dev)
+{
+ int err;
+
+ err = netdev_master_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_master_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_master_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);
@@ -3258,36 +3278,32 @@ 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 21b68e5..0d282d2 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -258,6 +258,9 @@ 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
*
@@ -280,11 +283,9 @@ 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)
@@ -360,10 +361,9 @@ 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 ae612f4..095bbc7 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -2003,6 +2003,7 @@ 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,
--
1.8.0
^ permalink raw reply related
* [patch net-next 14/15] net: remove no longer used netdev_set_bond_master() and netdev_set_master()
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 52d1146..c3723ef 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -856,8 +856,7 @@ 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.
*
@@ -2648,9 +2647,6 @@ extern int netdev_master_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 1af3141..ccca4bf 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4824,69 +4824,6 @@ 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_master_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;
--
1.8.0
^ permalink raw reply related
* [patch net-next 15/15] net: kill dev->master
From: Jiri Pirko @ 2012-12-30 11:58 UTC (permalink / raw)
To: netdev
Cc: davem, edumazet, bhutchings, faisal.latif, shemminger, fbl,
roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-1-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 c3723ef..009bad3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1167,10 +1167,6 @@ 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() */
--
1.8.0
^ permalink raw reply related
* Re: [PATCH] net: fix checking boundary of valid vlan id
From: Florian Westphal @ 2012-12-30 12:25 UTC (permalink / raw)
To: akong; +Cc: netdev, davem
In-Reply-To: <1356848931-22193-1-git-send-email-akong@redhat.com>
akong@redhat.com <akong@redhat.com> wrote:
> From: Amos Kong <akong@redhat.com>
>
> 4096 is not a valid vlan id.
True.
But why shouldn't users be allowed to check for frames
with reserved value?
^ permalink raw reply
* Re: [PATCH net-next] team: implement carrier change
From: Jiri Pirko @ 2012-12-30 13:06 UTC (permalink / raw)
To: Michał Mirosław; +Cc: netdev, Flavio Leitner
In-Reply-To: <CAHXqBFKtm6y3V-OepX9NnDP1GT28J=tPRmEtmR68wACPT9Pq8g@mail.gmail.com>
30. 12. 2012 v 14:03, Michał Mirosław <mirqus@gmail.com>:
> 2012/12/30 Flavio Leitner <fbl@redhat.com>:
>> The user space teamd daemon may need to control the
>> master's carrier state depending on the selected mode.
> [...]
>> +static int team_change_carrier(struct net_device *dev, bool new_carrier)
>> +{
>> + if (new_carrier)
>> + netif_carrier_on(dev);
>> + else
>> + netif_carrier_off(dev);
>> + return 0;
>> +}
>> +
> [...]
>> + .ndo_change_carrier = team_change_carrier,
>
> I wonder what else you might put in the .ndo_change_carrier
> implementation? It's already the same for dummy device and tunnels
> would get more copies. Maybe it would be enough to allocate a bit in
> netdev->priv_flags and use that in dev_change_carrier() instead?
>
Flavio has a patch adding some more code here prepared already. Will be posted soon.
> Best Regards,
> Michał Mirosław
^ permalink raw reply
* Re: [PATCH net-next] team: implement carrier change
From: Michał Mirosław @ 2012-12-30 13:03 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, Flavio Leitner
In-Reply-To: <1356831061-1483-1-git-send-email-fbl@redhat.com>
2012/12/30 Flavio Leitner <fbl@redhat.com>:
> The user space teamd daemon may need to control the
> master's carrier state depending on the selected mode.
[...]
> +static int team_change_carrier(struct net_device *dev, bool new_carrier)
> +{
> + if (new_carrier)
> + netif_carrier_on(dev);
> + else
> + netif_carrier_off(dev);
> + return 0;
> +}
> +
[...]
> + .ndo_change_carrier = team_change_carrier,
I wonder what else you might put in the .ndo_change_carrier
implementation? It's already the same for dummy device and tunnels
would get more copies. Maybe it would be enough to allocate a bit in
netdev->priv_flags and use that in dev_change_carrier() instead?
Best Regards,
Michał Mirosław
^ permalink raw reply
* Re: [patch net-next 01/15] net: introduce upper device lists
From: Ben Greear @ 2012-12-30 18:00 UTC (permalink / raw)
To: Jiri Pirko
Cc: netdev, davem, edumazet, bhutchings, faisal.latif, shemminger,
fbl, roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-2-git-send-email-jiri@resnulli.us>
On 12/30/2012 03:58 AM, Jiri Pirko wrote:
> This lists are supposed to serve for storing pointers to all upper devices.
> Eventually it will replace dev->master pointer which is used for
> bonding, bridge, team but it cannot be used for vlan, macvlan where
> there might be multiple upper present. In case the upper link is
> replacement for dev->master, it is marked with "master" flag.
This is confusing and full of grammatical errors. For instance, the 'it'
in "team but it cannot be used"...you are talking about the list this
patch introduces, or the old dev->master?
>
> New upper device list resolves this limitation. Also, the information
> stored in lists is used for preventing looping setups like
> "bond->somethingelse->samebond"
>
> Signed-off-by: Jiri Pirko <jiri@resnulli.us>
> ---
> include/linux/netdevice.h | 14 +++
> net/core/dev.c | 237 +++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 247 insertions(+), 4 deletions(-)
>
> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
> index 6835b58..52d1146 100644
> --- a/include/linux/netdevice.h
> +++ b/include/linux/netdevice.h
> @@ -1172,6 +1172,8 @@ struct net_device {
> * which this device is member of.
> */
>
> + struct list_head upper_dev_list; /* List of upper devices */
> +
Unless I mis-understand, this is often contains 'lower' devices instead,
and could contain a mixture.
Maybe this could use a rename, or at least a lot more comments?
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [patch net-next 04/15] rtnetlink: remove usage of dev->master
From: Ben Greear @ 2012-12-30 18:00 UTC (permalink / raw)
To: Jiri Pirko
Cc: netdev, davem, edumazet, bhutchings, faisal.latif, shemminger,
fbl, roland, sean.hefty, hal.rosenstock, fubar, andy, divy,
jitendra.kalsaria, sony.chacko, linux-driver, kaber, ursula.braun,
blaschka, schwidefsky, heiko.carstens, ebiederm, joe, amwang,
nhorman, john.r.fastabend, pablo
In-Reply-To: <1356868702-8144-5-git-send-email-jiri@resnulli.us>
On 12/30/2012 03:58 AM, Jiri Pirko wrote:
> Signed-off-by: Jiri Pirko <jiri@resnulli.us>
> ---
> net/core/rtnetlink.c | 69 ++++++++++++++++++++++++++++------------------------
> 1 file changed, 37 insertions(+), 32 deletions(-)
>
> diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
> index 2ef7a56..ae612f4 100644
> --- a/net/core/rtnetlink.c
> +++ b/net/core/rtnetlink.c
> @@ -880,6 +880,7 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
> const struct rtnl_link_stats64 *stats;
> struct nlattr *attr, *af_spec;
> struct rtnl_af_ops *af_ops;
> + struct net_device *upper_dev = netdev_master_upper_dev_get(dev);
>
> ASSERT_RTNL();
> nlh = nlmsg_put(skb, pid, seq, type, sizeof(*ifm), flags);
> @@ -908,8 +909,8 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
> #endif
> (dev->ifindex != dev->iflink &&
> nla_put_u32(skb, IFLA_LINK, dev->iflink)) ||
> - (dev->master &&
> - nla_put_u32(skb, IFLA_MASTER, dev->master->ifindex)) ||
> + (upper_dev &&
> + nla_put_u32(skb, IFLA_MASTER, upper_dev->ifindex)) ||
This seems confusing to me. You are calling the new variable 'upper_dev',
but it seems you are actually using it as a lower_dev???
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* [PATCH net-next] team: update master carrier state
From: Flavio Leitner @ 2012-12-30 18:27 UTC (permalink / raw)
To: netdev; +Cc: Jiri Pirko, Flavio Leitner
Update master's carrier state when there is any
change with its ports.
Signed-off-by: Flavio Leitner <fbl@redhat.com>
---
drivers/net/team/team.c | 24 ++++++++++++++++++++++--
1 file changed, 22 insertions(+), 2 deletions(-)
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index 918a901..78c7d87 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -1400,13 +1400,11 @@ static void team_destructor(struct net_device *dev)
static int team_open(struct net_device *dev)
{
- netif_carrier_on(dev);
return 0;
}
static int team_close(struct net_device *dev)
{
- netif_carrier_off(dev);
return 0;
}
@@ -2560,21 +2558,43 @@ send_event:
}
+static void __team_carrier_check(struct team *team)
+{
+ struct team_port *port;
+ bool team_linkup;
+
+ team_linkup = false;
+ list_for_each_entry(port, &team->port_list, list) {
+ if (port->linkup) {
+ team_linkup = true;
+ break;
+ }
+ }
+
+ if (team_linkup)
+ netif_carrier_on(team->dev);
+ else
+ netif_carrier_off(team->dev);
+}
+
static void __team_port_change_check(struct team_port *port, bool linkup)
{
if (port->state.linkup != linkup)
__team_port_change_send(port, linkup);
+ __team_carrier_check(port->team);
}
static void __team_port_change_port_added(struct team_port *port, bool linkup)
{
__team_port_change_send(port, linkup);
+ __team_carrier_check(port->team);
}
static void __team_port_change_port_removed(struct team_port *port)
{
port->removed = true;
__team_port_change_send(port, false);
+ __team_carrier_check(port->team);
}
static void team_port_change_check(struct team_port *port, bool linkup)
--
1.8.0.1
^ permalink raw reply related
* batman-adv: Unable to add interface in LXC
From: Pau Koning @ 2012-12-30 19:34 UTC (permalink / raw)
To: b.a.t.m.a.n-ZwoEplunGu2X36UT3dwllkB+6BGkLq7r
Cc: netdev-u79uwXL29TY76Z2rM5mHXA
I am running LXC (SID) under Debian SID and the current git version of
batman-adv (v3.7-rc7-1325-gaf5d4f7) +batctl (v2012.4.0-30-ga395164).
But it fails to add any of my interfaces and non-batman-adv interfaces
can be created without problems
# ip link
13: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
state UP mode DEFAULT qlen 1000
link/ether 00:ff:aa:00:00:01 brd ff:ff:ff:ff:ff:ff
15: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
mode DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
# batctl if add eth0
Error - can't open file '/sys/class/net/eth0/batman_adv/mesh_iface':
Read-only file system
# ls -l /sys/class/net/eth0/batman_adv/mesh_iface
-rw-r--r-- 1 root root 4096 Dec 30 18:26
/sys/class/net/eth0/batman_adv/mesh_iface
# id
uid=0(root) gid=0(root) groups=0(root)
# ip link add dev br0 type bridge
# ip link set dev eth0 master br0
# ip link
18: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
master br0 state UP mode DEFAULT qlen 1000
link/ether 00:ff:aa:00:00:01 brd ff:ff:ff:ff:ff:ff
20: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
mode DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
21: br0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT
link/ether 00:ff:aa:00:00:01 brd ff:ff:ff:ff:ff:ff
^ permalink raw reply
* [PATCH v3 net-next 00/22] bnx2x: support SR-IOV
From: Ariel Elior @ 2012-12-30 19:57 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Ariel Elior
Hi Dave,
Changes for v2:
-Remove redundant empty lines
-Remove redundant 'inline'
Changes for v3:
-Really remove all redundant empty lines
-Remove __dev* attributes from series
Changes for v4:
-Rearrange comments to be indented according to basic block
-Remove more redundant empty lines...
-use dev_info() rather than pr_info()
-add termnating newlines to all printouts (thanks Joe)
-add empty line between functional local variable declarations and actual code
This patch series adds support for SR-IOV in the bnx2x driver.
In bnx2x SR-IOV scheme the same bnx2x driver drives both VFs and PFs (single
binary).
The bulk of the communication between the VF drivers and the PF driver is done
via the VF <-> PF channel, a hardware-based communications channel with TLV
messages. The TLVs are designed to support different versions of VF drivers
(from multiple VMs) communicating with the same PF driver.
Patches:
01-03 - Probing and removing a VF driver
Includes sending the 'acquire' and 'release' messages on the VF PF channel.
Here the VF PF channel infrastructure is added, including the allocations of
the mailboxes and the definition of thin API structures (which will fill out
with more content as the series progresses).
04-07 - Loading/Unloading a VF driver
Includes refactoring of the driver load code and differentiating the VF flow
from the PF flow. This also includes requests from the VF for the PF to open
a queue in the HW on its behalf, configure the device with macs/vlans/rxmode
data, etc. Likewise the unload flow has been modified for the PF to undo these
configurations when the VF indicates it is going down.
08 - Modify fastpath flows for VFs
VFs have almost identical behavior in fastpath to PFs. In this patch the
VFs prepare transmit transaction for tx-switching, and the code for acking a
fastpath interrupt has been reorganized to allow a VF or PF to preconfigure the
offset of the interrupt's location in the BAR (as they have BARs with different
mapping) so as not to do so in fastpath.
09-10 - Probe and Load a PF driver with SR-IOV
The PF driver allocates and initializes the VF database to manage and keep track
of its VFs, their resources, queues, etc.
11-19 - The PF side of the VF <-> PF channel requests
Here is the implementation on the PF's side of the requests submitted in patches
01 through 07 by the VF. Patch 14 adds support for statistics collection by the
PF for all of its VFs (stats are DMAEed directly to VM GPA memory).
20 - Support for VF function level reset
When FLR indication is received for VFs, the PF reclaims all the resources
allocated for these VFs (interrupts, queues) and releases allocation it
performed for the FLRed VFs (it does so by consulting the VF database).
21 - Bulletin Board interface
This patch adds the PF <-> VF Bulletin Board interface. This interface is a
simple interface where the PF can be the initiator, and indicate to a VF that
it has a new MAC for it to use. In this interface each post "covers" any
previous posts (hence the name).
22 - Add the VF device ids and enable feature
In this patch we add the VF device ids of the various devices driven by bnx2x.
Here we also add the calls to "pci_enable_sriov" and "pci_disable_sriov".
Important: In this patch series we have laid the ground work for interfacing
with the infrastructure for dynamically controlling the number of VFs of a
physical device.
Meanwhile, in patch 9 this code is added to bnx2x_init_one():
--snip snip--
rc = bnx2x_iov_init_one(bp, int_mode, 0/*num vfs*/);
--snip snip--
recompiling the bnx2x with a number of vfs which is not 0 will enable SR-IOV
with that number of VFs.
Please consider applying these patches.
^ permalink raw reply
* [PATCH net-next v4 03/22] bnx2x: Add to VF <-> PF channel the release request
From: Ariel Elior @ 2012-12-30 19:57 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Ariel Elior, Eilon Greenstein
In-Reply-To: <1356897442-3063-1-git-send-email-ariele@broadcom.com>
VF driver uses this request when removed. The PF driver
reclaims all resources allocated for that VF at this
time.
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 1 +
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 43 ++++++++++++++++++++++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h | 15 ++++++++
3 files changed, 59 insertions(+), 0 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index eebc01d..4dc754b 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -2220,6 +2220,7 @@ static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms,
int bnx2x_get_vf_id(struct bnx2x *bp, u32 *vf_id);
int bnx2x_send_msg2pf(struct bnx2x *bp, u8 *done, dma_addr_t msg_mapping);
int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count);
+int bnx2x_vfpf_release(struct bnx2x *bp);
/* Congestion management fairness mode */
#define CMNG_FNS_NONE 0
#define CMNG_FNS_MINMAX 1
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index e53f851..3dd5b26 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -12363,6 +12363,9 @@ static void bnx2x_remove_one(struct pci_dev *pdev)
/* Make sure RESET task is not scheduled before continuing */
cancel_delayed_work_sync(&bp->sp_rtnl_task);
+ /* send message via vfpf channel to release the resources of this vf */
+ if (IS_VF(bp))
+ bnx2x_vfpf_release(bp);
if (bp->regview)
iounmap(bp->regview);
@@ -13343,3 +13346,43 @@ int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count)
return 0;
}
+
+int bnx2x_vfpf_release(struct bnx2x *bp)
+{
+ struct vfpf_release_tlv *req = &bp->vf2pf_mbox->req.release;
+ struct pfvf_general_resp_tlv *resp = &bp->vf2pf_mbox->resp.general_resp;
+ u32 rc = 0, vf_id;
+
+ /* clear mailbox and prep first tlv */
+ bnx2x_vfpf_prep(bp, &req->first_tlv, CHANNEL_TLV_RELEASE, sizeof(*req));
+
+ if (bnx2x_get_vf_id(bp, &vf_id))
+ return -EAGAIN;
+
+ req->vf_id = vf_id;
+
+ /* add list termination tlv */
+ bnx2x_add_tlv(bp, req, req->first_tlv.tl.length, CHANNEL_TLV_LIST_END,
+ sizeof(struct channel_list_end_tlv));
+
+ /* output tlvs list */
+ bnx2x_dp_tlv_list(bp, req);
+
+ /* send release request */
+ rc = bnx2x_send_msg2pf(bp, &resp->hdr.status, bp->vf2pf_mbox_mapping);
+
+ if (rc)
+ /* PF timeout */
+ return rc;
+ if (resp->hdr.status == PFVF_STATUS_SUCCESS) {
+ /* PF released us */
+ DP(BNX2X_MSG_SP, "vf released\n");
+ } else {
+ /* PF reports error */
+ BNX2X_ERR("PF failed our release request - are we out of sync? response status: %d\n",
+ resp->hdr.status);
+ return -EAGAIN;
+ }
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
index 728491f..561c656 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
@@ -68,6 +68,11 @@ struct pfvf_tlv {
u8 padding[3];
};
+/* response tlv used for most tlvs */
+struct pfvf_general_resp_tlv {
+ struct pfvf_tlv hdr;
+};
+
/* used to terminate and pad a tlv list */
struct channel_list_end_tlv {
struct channel_tlv tl;
@@ -125,6 +130,13 @@ struct pfvf_acquire_resp_tlv {
} resc;
};
+/* release the VF's acquired resources */
+struct vfpf_release_tlv {
+ struct vfpf_first_tlv first_tlv;
+ u16 vf_id;
+ u8 padding[2];
+};
+
struct tlv_buffer_size {
u8 tlv_buffer[TLV_BUFFER_SIZE];
};
@@ -132,11 +144,13 @@ struct tlv_buffer_size {
union vfpf_tlvs {
struct vfpf_first_tlv first_tlv;
struct vfpf_acquire_tlv acquire;
+ struct vfpf_release_tlv release;
struct channel_list_end_tlv list_end;
struct tlv_buffer_size tlv_buf_size;
};
union pfvf_tlvs {
+ struct pfvf_general_resp_tlv general_resp;
struct pfvf_acquire_resp_tlv acquire_resp;
struct channel_list_end_tlv list_end;
struct tlv_buffer_size tlv_buf_size;
@@ -147,6 +161,7 @@ union pfvf_tlvs {
enum channel_tlvs {
CHANNEL_TLV_NONE,
CHANNEL_TLV_ACQUIRE,
+ CHANNEL_TLV_RELEASE,
CHANNEL_TLV_LIST_END,
CHANNEL_TLV_MAX
};
--
1.7.9.GIT
^ permalink raw reply related
* [PATCH net-next v4 02/22] bnx2x: VF <-> PF channel 'acquire' at vf probe
From: Ariel Elior @ 2012-12-30 19:57 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Ariel Elior, Eilon Greenstein
In-Reply-To: <1356897442-3063-1-git-send-email-ariele@broadcom.com>
Add the 'acquire' request to VF <-> PF channel and use it at
VF probe. In the acquire request the VF driver lists the resources
it would like to have. In the response the PF either ratifies the
request, or denies it and supplies the maximum values supported.
The VF may then attempt another acquire request.
This patch adds the bnx2x_vfpf.c file which contains the
implementation of the VF to PF hardware channel.
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2x/Makefile | 2 +-
drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 13 ++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 185 +++++++++++++++++++++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h | 6 +
drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h | 5 +
drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c | 80 +++++++++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h | 117 +++++++++++++
7 files changed, 407 insertions(+), 1 deletions(-)
create mode 100644 drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
diff --git a/drivers/net/ethernet/broadcom/bnx2x/Makefile b/drivers/net/ethernet/broadcom/bnx2x/Makefile
index 48fbdd4..d862ea6 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/Makefile
+++ b/drivers/net/ethernet/broadcom/bnx2x/Makefile
@@ -4,4 +4,4 @@
obj-$(CONFIG_BNX2X) += bnx2x.o
-bnx2x-objs := bnx2x_main.o bnx2x_link.o bnx2x_cmn.o bnx2x_ethtool.o bnx2x_stats.o bnx2x_dcb.o bnx2x_sp.o
+bnx2x-objs := bnx2x_main.o bnx2x_link.o bnx2x_cmn.o bnx2x_ethtool.o bnx2x_stats.o bnx2x_dcb.o bnx2x_sp.o bnx2x_vfpf.o
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index ff5da33..eebc01d 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -48,6 +48,7 @@
#include "bnx2x_sp.h"
#include "bnx2x_dcb.h"
#include "bnx2x_stats.h"
+#include "bnx2x_vfpf.h"
enum bnx2x_int_mode {
BNX2X_INT_MODE_MSIX,
@@ -1244,6 +1245,9 @@ struct bnx2x {
struct bnx2x_vf_mbx_msg *vf2pf_mbox;
dma_addr_t vf2pf_mbox_mapping;
+ /* we set aside a copy of the acquire response */
+ struct pfvf_acquire_resp_tlv acquire_resp;
+
struct net_device *dev;
struct pci_dev *pdev;
@@ -2207,6 +2211,15 @@ static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms,
#define BNX2X_VPD_LEN 128
#define VENDOR_ID_LEN 4
+#define VF_ACQUIRE_THRESH 3
+#define VF_ACQUIRE_MAC_FILTERS 1
+#define VF_ACQUIRE_MC_FILTERS 10
+
+#define GOOD_ME_REG(me_reg) (((me_reg) & ME_REG_VF_VALID) && \
+ (!((me_reg) & ME_REG_VF_ERR)))
+int bnx2x_get_vf_id(struct bnx2x *bp, u32 *vf_id);
+int bnx2x_send_msg2pf(struct bnx2x *bp, u8 *done, dma_addr_t msg_mapping);
+int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count);
/* Congestion management fairness mode */
#define CMNG_FNS_NONE 0
#define CMNG_FNS_MINMAX 1
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index bb59004..e53f851 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -12234,6 +12234,12 @@ static int bnx2x_init_one(struct pci_dev *pdev,
goto init_one_exit;
}
+ if (IS_VF(bp)) {
+ rc = bnx2x_vfpf_acquire(bp, tx_count, rx_count);
+ if (rc)
+ goto init_one_exit;
+ }
+
/* calc qm_cid_count */
bp->qm_cid_count = bnx2x_set_qm_cid_count(bp);
BNX2X_DEV_INFO("qm_cid_count %d\n", bp->qm_cid_count);
@@ -13157,4 +13163,183 @@ struct cnic_eth_dev *bnx2x_cnic_probe(struct net_device *dev)
return cp;
}
+int bnx2x_send_msg2pf(struct bnx2x *bp, u8 *done, dma_addr_t msg_mapping)
+{
+ struct cstorm_vf_zone_data __iomem *zone_data =
+ REG_ADDR(bp, PXP_VF_ADDR_CSDM_GLOBAL_START);
+ int tout = 600, interval = 100; /* wait for 60 seconds */
+
+ if (*done) {
+ BNX2X_ERR("done was non zero before message to pf was sent\n");
+ WARN_ON(true);
+ return -EINVAL;
+ }
+
+ /* Write message address */
+ writel(U64_LO(msg_mapping),
+ &zone_data->non_trigger.vf_pf_channel.msg_addr_lo);
+ writel(U64_HI(msg_mapping),
+ &zone_data->non_trigger.vf_pf_channel.msg_addr_hi);
+
+ /* make sure the address is written before FW accesses it */
+ wmb();
+
+ /* Trigger the PF FW */
+ writeb(1, &zone_data->trigger.vf_pf_channel.addr_valid);
+
+ /* Wait for PF to complete */
+ while ((tout >= 0) && (!*done)) {
+ msleep(interval);
+ tout -= 1;
+
+ /* progress indicator - HV can take its own sweet time in
+ * answering VFs...
+ */
+ DP_CONT(BNX2X_MSG_IOV, ".");
+ }
+
+ if (!*done) {
+ BNX2X_ERR("PF response has timed out\n");
+ return -EAGAIN;
+ }
+ DP(BNX2X_MSG_SP, "Got a response from PF\n");
+ return 0;
+}
+
+int bnx2x_get_vf_id(struct bnx2x *bp, u32 *vf_id)
+{
+ u32 me_reg;
+ int tout = 10, interval = 100; /* Wait for 1 sec */
+
+ do {
+ /* pxp traps vf read of doorbells and returns me reg value */
+ me_reg = readl(bp->doorbells);
+ if (GOOD_ME_REG(me_reg))
+ break;
+
+ msleep(interval);
+
+ BNX2X_ERR("Invalid ME register value: 0x%08x\n. Is pf driver up?",
+ me_reg);
+ } while (tout-- > 0);
+
+ if (!GOOD_ME_REG(me_reg)) {
+ BNX2X_ERR("Invalid ME register value: 0x%08x\n", me_reg);
+ return -EINVAL;
+ }
+
+ BNX2X_ERR("valid ME register value: 0x%08x\n", me_reg);
+
+ *vf_id = (me_reg & ME_REG_VF_NUM_MASK) >> ME_REG_VF_NUM_SHIFT;
+ return 0;
+}
+
+int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count)
+{
+ int rc = 0, attempts = 0;
+ struct vfpf_acquire_tlv *req = &bp->vf2pf_mbox->req.acquire;
+ struct pfvf_acquire_resp_tlv *resp = &bp->vf2pf_mbox->resp.acquire_resp;
+ u32 vf_id;
+ bool resources_acquired = false;
+
+ /* clear mailbox and prep first tlv */
+ bnx2x_vfpf_prep(bp, &req->first_tlv, CHANNEL_TLV_ACQUIRE, sizeof(*req));
+
+ if (bnx2x_get_vf_id(bp, &vf_id))
+ return -EAGAIN;
+
+ req->vfdev_info.vf_id = vf_id;
+ req->vfdev_info.vf_os = 0;
+
+ req->resc_request.num_rxqs = rx_count;
+ req->resc_request.num_txqs = tx_count;
+ req->resc_request.num_sbs = bp->igu_sb_cnt;
+ req->resc_request.num_mac_filters = VF_ACQUIRE_MAC_FILTERS;
+ req->resc_request.num_mc_filters = VF_ACQUIRE_MC_FILTERS;
+
+ /* add list termination tlv */
+ bnx2x_add_tlv(bp, req, req->first_tlv.tl.length, CHANNEL_TLV_LIST_END,
+ sizeof(struct channel_list_end_tlv));
+
+ /* output tlvs list */
+ bnx2x_dp_tlv_list(bp, req);
+
+ while (!resources_acquired) {
+ DP(BNX2X_MSG_SP, "attempting to acquire resources\n");
+
+ /* send acquire request */
+ rc = bnx2x_send_msg2pf(bp,
+ &resp->hdr.status,
+ bp->vf2pf_mbox_mapping);
+
+ /* PF timeout */
+ if (rc)
+ return rc;
+
+ /* copy acquire response from buffer to bp */
+ memcpy(&bp->acquire_resp, resp, sizeof(bp->acquire_resp));
+
+ attempts++;
+
+ /* test whether the PF accepted our request. If not, humble the
+ * the request and try again.
+ */
+ if (bp->acquire_resp.hdr.status == PFVF_STATUS_SUCCESS) {
+ DP(BNX2X_MSG_SP, "resources acquired\n");
+ resources_acquired = true;
+ } else if (bp->acquire_resp.hdr.status ==
+ PFVF_STATUS_NO_RESOURCE &&
+ attempts < VF_ACQUIRE_THRESH) {
+ DP(BNX2X_MSG_SP,
+ "PF unwilling to fulfill resource request. Try PF recommended amount\n");
+
+ /* humble our request */
+ req->resc_request.num_txqs =
+ bp->acquire_resp.resc.num_txqs;
+ req->resc_request.num_rxqs =
+ bp->acquire_resp.resc.num_rxqs;
+ req->resc_request.num_sbs =
+ bp->acquire_resp.resc.num_sbs;
+ req->resc_request.num_mac_filters =
+ bp->acquire_resp.resc.num_mac_filters;
+ req->resc_request.num_vlan_filters =
+ bp->acquire_resp.resc.num_vlan_filters;
+ req->resc_request.num_mc_filters =
+ bp->acquire_resp.resc.num_mc_filters;
+
+ /* Clear response buffer */
+ memset(&bp->vf2pf_mbox->resp, 0,
+ sizeof(union pfvf_tlvs));
+ } else {
+ /* PF reports error */
+ BNX2X_ERR("Failed to get the requested amount of resources: %d. Breaking...\n",
+ bp->acquire_resp.hdr.status);
+ return -EAGAIN;
+ }
+ }
+
+ /* get HW info */
+ bp->common.chip_id |= (bp->acquire_resp.pfdev_info.chip_num & 0xffff);
+ bp->link_params.chip_id = bp->common.chip_id;
+ bp->db_size = bp->acquire_resp.pfdev_info.db_size;
+ bp->common.int_block = INT_BLOCK_IGU;
+ bp->common.chip_port_mode = CHIP_2_PORT_MODE;
+ bp->igu_dsb_id = -1;
+ bp->mf_ov = 0;
+ bp->mf_mode = 0;
+ bp->common.flash_size = 0;
+ bp->flags |=
+ NO_WOL_FLAG | NO_ISCSI_OOO_FLAG | NO_ISCSI_FLAG | NO_FCOE_FLAG;
+ bp->igu_sb_cnt = 1;
+ bp->igu_base_sb = bp->acquire_resp.resc.hw_sbs[0].hw_sb_id;
+ strlcpy(bp->fw_ver, bp->acquire_resp.pfdev_info.fw_ver,
+ sizeof(bp->fw_ver));
+
+ if (is_valid_ether_addr(bp->acquire_resp.resc.current_mac_addr))
+ memcpy(bp->dev->dev_addr,
+ bp->acquire_resp.resc.current_mac_addr,
+ ETH_ALEN);
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index 463a984..c302de4 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -6558,6 +6558,12 @@
#define PXP_VF_ADDR_IGU_SIZE 0x3000
#define PXP_VF_ADDR_IGU_END\
((PXP_VF_ADDR_IGU_START) + (PXP_VF_ADDR_IGU_SIZE) - 1)
+
+#define PXP_VF_ADDR_CSDM_GLOBAL_START 0x7600
+#define PXP_VF_ADDR_CSDM_GLOBAL_SIZE (PXP_ADDR_REG_SIZE)
+#define PXP_VF_ADDR_CSDM_GLOBAL_END\
+ ((PXP_VF_ADDR_CSDM_GLOBAL_START) + (PXP_VF_ADDR_CSDM_GLOBAL_SIZE) - 1)
+
#define PXP_VF_ADDR_DB_START 0x7c00
#define PXP_VF_ADDR_DB_SIZE 0x200
#define PXP_VF_ADDR_DB_END\
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
index 1b14745..6d0df33 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
@@ -24,4 +24,9 @@ struct bnx2x_vf_mbx_msg {
union pfvf_tlvs resp;
};
+void bnx2x_add_tlv(struct bnx2x *bp, void *tlvs_list, u16 offset, u16 type,
+ u16 length);
+void bnx2x_vfpf_prep(struct bnx2x *bp, struct vfpf_first_tlv *first_tlv,
+ u16 type, u16 length);
+void bnx2x_dp_tlv_list(struct bnx2x *bp, void *tlvs_list);
#endif /* bnx2x_sriov.h */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
new file mode 100644
index 0000000..55dd6a9
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
@@ -0,0 +1,80 @@
+/* bnx2x_vfpf.c: Broadcom Everest network driver.
+ *
+ * Copyright 2009-2012 Broadcom Corporation
+ *
+ * Unless you and Broadcom execute a separate written software license
+ * agreement governing use of this software, this software is licensed to you
+ * under the terms of the GNU General Public License version 2, available
+ * at http://www.gnu.org/licenses/old-licenses/gpl-2.0.html (the "GPL").
+ *
+ * Notwithstanding the above, under no circumstances may you combine this
+ * software in any way with any other Broadcom software provided under a
+ * license other than the GPL, without Broadcom's express prior written
+ * consent.
+ *
+ * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Written by: Shmulik Ravid <shmulikr@broadcom.com>
+ * Ariel Elior <ariele@broadcom.com>
+ */
+
+#include "bnx2x.h"
+#include "bnx2x_sriov.h"
+
+/* place a given tlv on the tlv buffer at a given offset */
+void bnx2x_add_tlv(struct bnx2x *bp, void *tlvs_list, u16 offset, u16 type,
+ u16 length)
+{
+ struct channel_tlv *tl =
+ (struct channel_tlv *)(tlvs_list + offset);
+
+ tl->type = type;
+ tl->length = length;
+}
+
+/* Clear the mailbox and init the header of the first tlv */
+void bnx2x_vfpf_prep(struct bnx2x *bp, struct vfpf_first_tlv *first_tlv,
+ u16 type, u16 length)
+{
+ DP(BNX2X_MSG_IOV, "preparing to send %d tlv over vf pf channel\n",
+ type);
+
+ /* Clear mailbox */
+ memset(bp->vf2pf_mbox, 0, sizeof(struct bnx2x_vf_mbx_msg));
+
+ /* init type and length */
+ bnx2x_add_tlv(bp, &first_tlv->tl, 0, type, length);
+
+ /* init first tlv header */
+ first_tlv->resp_msg_offset = sizeof(bp->vf2pf_mbox->req);
+}
+
+/* list the types and lengths of the tlvs on the buffer */
+void bnx2x_dp_tlv_list(struct bnx2x *bp, void *tlvs_list)
+{
+ int i = 1;
+ struct channel_tlv *tlv = (struct channel_tlv *)tlvs_list;
+
+ while (tlv->type != CHANNEL_TLV_LIST_END) {
+ /* output tlv */
+ DP(BNX2X_MSG_IOV, "TLV number %d: type %d, length %d\n", i,
+ tlv->type, tlv->length);
+
+ /* advance to next tlv */
+ tlvs_list += tlv->length;
+
+ /* cast general tlv list pointer to channel tlv header*/
+ tlv = (struct channel_tlv *)tlvs_list;
+
+ i++;
+
+ /* break condition for this loop */
+ if (i > MAX_TLVS_IN_LIST) {
+ WARN(true, "corrupt tlvs");
+ return;
+ }
+ }
+
+ /* output last tlv */
+ DP(BNX2X_MSG_IOV, "TLV number %d: type %d, length %d\n", i,
+ tlv->type, tlv->length);
+}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
index bb37675..728491f 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
@@ -18,20 +18,137 @@
#ifndef VF_PF_IF_H
#define VF_PF_IF_H
+/* Common definitions for all HVs */
+struct vf_pf_resc_request {
+ u8 num_rxqs;
+ u8 num_txqs;
+ u8 num_sbs;
+ u8 num_mac_filters;
+ u8 num_vlan_filters;
+ u8 num_mc_filters; /* No limit so superfluous */
+};
+
+struct hw_sb_info {
+ u8 hw_sb_id; /* aka absolute igu id, used to ack the sb */
+ u8 sb_qid; /* used to update DHC for sb */
+};
+
/* HW VF-PF channel definitions
* A.K.A VF-PF mailbox
*/
#define TLV_BUFFER_SIZE 1024
+enum {
+ PFVF_STATUS_WAITING = 0,
+ PFVF_STATUS_SUCCESS,
+ PFVF_STATUS_FAILURE,
+ PFVF_STATUS_NOT_SUPPORTED,
+ PFVF_STATUS_NO_RESOURCE
+};
+
+/* vf pf channel tlvs */
+/* general tlv header (used for both vf->pf request and pf->vf response) */
+struct channel_tlv {
+ u16 type;
+ u16 length;
+};
+
+/* header of first vf->pf tlv carries the offset used to calculate response
+ * buffer address
+ */
+struct vfpf_first_tlv {
+ struct channel_tlv tl;
+ u32 resp_msg_offset;
+};
+
+/* header of pf->vf tlvs, carries the status of handling the request */
+struct pfvf_tlv {
+ struct channel_tlv tl;
+ u8 status;
+ u8 padding[3];
+};
+
+/* used to terminate and pad a tlv list */
+struct channel_list_end_tlv {
+ struct channel_tlv tl;
+ u8 padding[4];
+};
+
+/* Acquire */
+struct vfpf_acquire_tlv {
+ struct vfpf_first_tlv first_tlv;
+
+ struct vf_pf_vfdev_info {
+ /* the following fields are for debug purposes */
+ u8 vf_id; /* ME register value */
+ u8 vf_os; /* e.g. Linux, W2K8 */
+ u8 padding[2];
+ } vfdev_info;
+
+ struct vf_pf_resc_request resc_request;
+
+ aligned_u64 bulletin_addr;
+};
+
+/* acquire response tlv - carries the allocated resources */
+struct pfvf_acquire_resp_tlv {
+ struct pfvf_tlv hdr;
+ struct pf_vf_pfdev_info {
+ u32 chip_num;
+ u32 pf_cap;
+#define PFVF_CAP_RSS 0x00000001
+#define PFVF_CAP_DHC 0x00000002
+#define PFVF_CAP_TPA 0x00000004
+ char fw_ver[32];
+ u16 db_size;
+ u8 indices_per_sb;
+ u8 padding;
+ } pfdev_info;
+ struct pf_vf_resc {
+ /* in case of status NO_RESOURCE in message hdr, pf will fill
+ * this struct with suggested amount of resources for next
+ * acquire request
+ */
+#define PFVF_MAX_QUEUES_PER_VF 16
+#define PFVF_MAX_SBS_PER_VF 16
+ struct hw_sb_info hw_sbs[PFVF_MAX_SBS_PER_VF];
+ u8 hw_qid[PFVF_MAX_QUEUES_PER_VF];
+ u8 num_rxqs;
+ u8 num_txqs;
+ u8 num_sbs;
+ u8 num_mac_filters;
+ u8 num_vlan_filters;
+ u8 num_mc_filters;
+ u8 permanent_mac_addr[ETH_ALEN];
+ u8 current_mac_addr[ETH_ALEN];
+ u8 padding[2];
+ } resc;
+};
+
struct tlv_buffer_size {
u8 tlv_buffer[TLV_BUFFER_SIZE];
};
union vfpf_tlvs {
+ struct vfpf_first_tlv first_tlv;
+ struct vfpf_acquire_tlv acquire;
+ struct channel_list_end_tlv list_end;
struct tlv_buffer_size tlv_buf_size;
};
union pfvf_tlvs {
+ struct pfvf_acquire_resp_tlv acquire_resp;
+ struct channel_list_end_tlv list_end;
struct tlv_buffer_size tlv_buf_size;
};
+
+#define MAX_TLVS_IN_LIST 50
+
+enum channel_tlvs {
+ CHANNEL_TLV_NONE,
+ CHANNEL_TLV_ACQUIRE,
+ CHANNEL_TLV_LIST_END,
+ CHANNEL_TLV_MAX
+};
+
#endif /* VF_PF_IF_H */
--
1.7.9.GIT
^ permalink raw reply related
* [PATCH net-next v4 01/22] bnx2x: Support probing and removing of VF device
From: Ariel Elior @ 2012-12-30 19:57 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Ariel Elior, Eilon Greenstein
In-Reply-To: <1356897442-3063-1-git-send-email-ariele@broadcom.com>
To support probing and removing of a bnx2x virtual function
the following were added:
1. add bnx2x_vfpf.h: defines the VF to PF channel
2. add bnx2x_sriov.h: header for bnx2x SR-IOV functionality
3. enumerate VF hw types (identify VFs)
4. if driving a VF, map VF bar
5. if driving a VF, allocate Vf to PF channel
6. refactor interrupt flows to include VF
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 21 +-
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 25 +-
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 2 +-
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 390 +++++++++++++--------
drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h | 9 +
drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h | 27 ++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h | 37 ++
7 files changed, 353 insertions(+), 158 deletions(-)
create mode 100644 drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
create mode 100644 drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index e8d4db1..ff5da33 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -49,6 +49,12 @@
#include "bnx2x_dcb.h"
#include "bnx2x_stats.h"
+enum bnx2x_int_mode {
+ BNX2X_INT_MODE_MSIX,
+ BNX2X_INT_MODE_INTX,
+ BNX2X_INT_MODE_MSI
+};
+
/* error/debug prints */
#define DRV_MODULE_NAME "bnx2x"
@@ -954,6 +960,9 @@ struct bnx2x_port {
extern struct workqueue_struct *bnx2x_wq;
#define BNX2X_MAX_NUM_OF_VFS 64
+#define BNX2X_VF_CID_WND 0
+#define BNX2X_CIDS_PER_VF (1 << BNX2X_VF_CID_WND)
+#define BNX2X_VF_CIDS (BNX2X_MAX_NUM_OF_VFS * BNX2X_CIDS_PER_VF)
#define BNX2X_VF_ID_INVALID 0xFF
/*
@@ -1231,6 +1240,10 @@ struct bnx2x {
(vn) * ((CHIP_IS_E1x(bp) || (CHIP_MODE_IS_4_PORT(bp))) ? 2 : 1))
#define BP_FW_MB_IDX(bp) BP_FW_MB_IDX_VN(bp, BP_VN(bp))
+ /* vf pf channel mailbox contains request and response buffers */
+ struct bnx2x_vf_mbx_msg *vf2pf_mbox;
+ dma_addr_t vf2pf_mbox_mapping;
+
struct net_device *dev;
struct pci_dev *pdev;
@@ -1318,8 +1331,6 @@ struct bnx2x {
#define DISABLE_MSI_FLAG (1 << 7)
#define TPA_ENABLE_FLAG (1 << 8)
#define NO_MCP_FLAG (1 << 9)
-
-#define BP_NOMCP(bp) (bp->flags & NO_MCP_FLAG)
#define GRO_ENABLE_FLAG (1 << 10)
#define MF_FUNC_DIS (1 << 11)
#define OWN_CNIC_IRQ (1 << 12)
@@ -1330,6 +1341,11 @@ struct bnx2x {
#define BC_SUPPORTS_FCOE_FEATURES (1 << 19)
#define USING_SINGLE_MSIX_FLAG (1 << 20)
#define BC_SUPPORTS_DCBX_MSG_NON_PMF (1 << 21)
+#define IS_VF_FLAG (1 << 22)
+
+#define BP_NOMCP(bp) ((bp)->flags & NO_MCP_FLAG)
+#define IS_VF(bp) ((bp)->flags & IS_VF_FLAG)
+#define IS_PF(bp) (!((bp)->flags & IS_VF_FLAG))
#define NO_ISCSI(bp) ((bp)->flags & NO_ISCSI_FLAG)
#define NO_ISCSI_OOO(bp) ((bp)->flags & NO_ISCSI_OOO_FLAG)
@@ -1432,6 +1448,7 @@ struct bnx2x {
u8 igu_sb_cnt;
u8 min_msix_vec_cnt;
+ u32 igu_base_addr;
dma_addr_t def_status_blk_mapping;
struct bnx2x_slowpath *slowpath;
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 01588b6..8687eb3 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -1439,12 +1439,15 @@ void bnx2x_free_irq(struct bnx2x *bp)
int bnx2x_enable_msix(struct bnx2x *bp)
{
- int msix_vec = 0, i, rc, req_cnt;
+ int msix_vec = 0, i, rc;
- bp->msix_table[msix_vec].entry = msix_vec;
- BNX2X_DEV_INFO("msix_table[0].entry = %d (slowpath)\n",
- bp->msix_table[0].entry);
- msix_vec++;
+ /* VFs don't have a default status block */
+ if (IS_PF(bp)) {
+ bp->msix_table[msix_vec].entry = msix_vec;
+ BNX2X_DEV_INFO("msix_table[0].entry = %d (slowpath)\n",
+ bp->msix_table[0].entry);
+ msix_vec++;
+ }
/* Cnic requires an msix vector for itself */
if (CNIC_SUPPORT(bp)) {
@@ -1462,9 +1465,10 @@ int bnx2x_enable_msix(struct bnx2x *bp)
msix_vec++;
}
- req_cnt = BNX2X_NUM_ETH_QUEUES(bp) + CNIC_SUPPORT(bp) + 1;
+ DP(BNX2X_MSG_SP, "about to request enable msix with %d vectors\n",
+ msix_vec);
- rc = pci_enable_msix(bp->pdev, &bp->msix_table[0], req_cnt);
+ rc = pci_enable_msix(bp->pdev, &bp->msix_table[0], msix_vec);
/*
* reconfigure number of tx/rx queues according to available
@@ -1472,7 +1476,7 @@ int bnx2x_enable_msix(struct bnx2x *bp)
*/
if (rc >= BNX2X_MIN_MSIX_VEC_CNT(bp)) {
/* how less vectors we will have? */
- int diff = req_cnt - rc;
+ int diff = msix_vec - rc;
BNX2X_DEV_INFO("Trying to use less MSI-X vectors: %d\n", rc);
@@ -3905,7 +3909,10 @@ int bnx2x_alloc_mem_bp(struct bnx2x *bp)
* The biggest MSI-X table we might need is as a maximum number of fast
* path IGU SBs plus default SB (for PF).
*/
- msix_table_size = bp->igu_sb_cnt + 1;
+ msix_table_size = bp->igu_sb_cnt;
+ if (IS_PF(bp))
+ msix_table_size++;
+ BNX2X_DEV_INFO("msix_table_size %d\n", msix_table_size);
/* fp array: RSS plus CNIC related L2 queues */
fp_array_size = BNX2X_MAX_RSS_COUNT(bp) + CNIC_SUPPORT(bp);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index 0991534..bca371e 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -863,7 +863,7 @@ static inline void bnx2x_del_all_napi(struct bnx2x *bp)
netif_napi_del(&bnx2x_fp(bp, i, napi));
}
-void bnx2x_set_int_mode(struct bnx2x *bp);
+int bnx2x_set_int_mode(struct bnx2x *bp);
static inline void bnx2x_disable_msi(struct bnx2x *bp)
{
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 940ef85..bb59004 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -59,6 +59,8 @@
#include "bnx2x_init.h"
#include "bnx2x_init_ops.h"
#include "bnx2x_cmn.h"
+#include "bnx2x_vfpf.h"
+#include "bnx2x_sriov.h"
#include "bnx2x_dcb.h"
#include "bnx2x_sp.h"
@@ -133,39 +135,49 @@ enum bnx2x_board_type {
BCM57711E,
BCM57712,
BCM57712_MF,
+ BCM57712_VF,
BCM57800,
BCM57800_MF,
+ BCM57800_VF,
BCM57810,
BCM57810_MF,
- BCM57840_O,
+ BCM57810_VF,
BCM57840_4_10,
BCM57840_2_20,
- BCM57840_MFO,
BCM57840_MF,
+ BCM57840_VF,
BCM57811,
- BCM57811_MF
+ BCM57811_MF,
+ BCM57840_O,
+ BCM57840_MFO,
+ BCM57811_VF
};
/* indexed by board_type, above */
static struct {
char *name;
} board_info[] = {
- { "Broadcom NetXtreme II BCM57710 10 Gigabit PCIe [Everest]" },
- { "Broadcom NetXtreme II BCM57711 10 Gigabit PCIe" },
- { "Broadcom NetXtreme II BCM57711E 10 Gigabit PCIe" },
- { "Broadcom NetXtreme II BCM57712 10 Gigabit Ethernet" },
- { "Broadcom NetXtreme II BCM57712 10 Gigabit Ethernet Multi Function" },
- { "Broadcom NetXtreme II BCM57800 10 Gigabit Ethernet" },
- { "Broadcom NetXtreme II BCM57800 10 Gigabit Ethernet Multi Function" },
- { "Broadcom NetXtreme II BCM57810 10 Gigabit Ethernet" },
- { "Broadcom NetXtreme II BCM57810 10 Gigabit Ethernet Multi Function" },
- { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet" },
- { "Broadcom NetXtreme II BCM57840 10 Gigabit Ethernet" },
- { "Broadcom NetXtreme II BCM57840 20 Gigabit Ethernet" },
- { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet Multi Function"},
- { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet Multi Function"},
- { "Broadcom NetXtreme II BCM57811 10 Gigabit Ethernet"},
- { "Broadcom NetXtreme II BCM57811 10 Gigabit Ethernet Multi Function"},
+ [BCM57710] = { "Broadcom NetXtreme II BCM57710 10 Gigabit PCIe [Everest]" },
+ [BCM57711] = { "Broadcom NetXtreme II BCM57711 10 Gigabit PCIe" },
+ [BCM57711E] = { "Broadcom NetXtreme II BCM57711E 10 Gigabit PCIe" },
+ [BCM57712] = { "Broadcom NetXtreme II BCM57712 10 Gigabit Ethernet" },
+ [BCM57712_MF] = { "Broadcom NetXtreme II BCM57712 10 Gigabit Ethernet Multi Function" },
+ [BCM57712_VF] = { "Broadcom NetXtreme II BCM57712 10 Gigabit Ethernet Virtual Function" },
+ [BCM57800] = { "Broadcom NetXtreme II BCM57800 10 Gigabit Ethernet" },
+ [BCM57800_MF] = { "Broadcom NetXtreme II BCM57800 10 Gigabit Ethernet Multi Function" },
+ [BCM57800_VF] = { "Broadcom NetXtreme II BCM57800 10 Gigabit Ethernet Virtual Function" },
+ [BCM57810] = { "Broadcom NetXtreme II BCM57810 10 Gigabit Ethernet" },
+ [BCM57810_MF] = { "Broadcom NetXtreme II BCM57810 10 Gigabit Ethernet Multi Function" },
+ [BCM57810_VF] = { "Broadcom NetXtreme II BCM57810 10 Gigabit Ethernet Virtual Function" },
+ [BCM57840_4_10] = { "Broadcom NetXtreme II BCM57840 10 Gigabit Ethernet" },
+ [BCM57840_2_20] = { "Broadcom NetXtreme II BCM57840 20 Gigabit Ethernet" },
+ [BCM57840_MF] = { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet Multi Function" },
+ [BCM57840_VF] = { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet Virtual Function" },
+ [BCM57811] = { "Broadcom NetXtreme II BCM57811 10 Gigabit Ethernet" },
+ [BCM57811_MF] = { "Broadcom NetXtreme II BCM57811 10 Gigabit Ethernet Multi Function" },
+ [BCM57840_O] = { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet" },
+ [BCM57840_MFO] = { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet Multi Function" },
+ [BCM57811_VF] = { "Broadcom NetXtreme II BCM57840 10/20 Gigabit Ethernet Virtual Function" }
};
#ifndef PCI_DEVICE_ID_NX2_57710
@@ -7792,41 +7804,49 @@ int bnx2x_setup_leading(struct bnx2x *bp)
*
* In case of MSI-X it will also try to enable MSI-X.
*/
-void bnx2x_set_int_mode(struct bnx2x *bp)
+int bnx2x_set_int_mode(struct bnx2x *bp)
{
+ int rc = 0;
+
+ if (IS_VF(bp) && int_mode != BNX2X_INT_MODE_MSIX)
+ return -EINVAL;
+
switch (int_mode) {
- case INT_MODE_MSI:
+ case BNX2X_INT_MODE_MSIX:
+ /* attempt to enable msix */
+ rc = bnx2x_enable_msix(bp);
+
+ /* msix attained */
+ if (!rc)
+ return 0;
+
+ /* vfs use only msix */
+ if (rc && IS_VF(bp))
+ return rc;
+
+ /* failed to enable multiple MSI-X */
+ BNX2X_DEV_INFO("Failed to enable multiple MSI-X (%d), set number of queues to %d\n",
+ bp->num_queues,
+ 1 + bp->num_cnic_queues);
+
+ /* falling through... */
+ case BNX2X_INT_MODE_MSI:
bnx2x_enable_msi(bp);
+
/* falling through... */
- case INT_MODE_INTx:
+ case BNX2X_INT_MODE_INTX:
bp->num_ethernet_queues = 1;
bp->num_queues = bp->num_ethernet_queues + bp->num_cnic_queues;
BNX2X_DEV_INFO("set number of queues to 1\n");
break;
default:
- /* if we can't use MSI-X we only need one fp,
- * so try to enable MSI-X with the requested number of fp's
- * and fallback to MSI or legacy INTx with one fp
- */
- if (bnx2x_enable_msix(bp) ||
- bp->flags & USING_SINGLE_MSIX_FLAG) {
- /* failed to enable multiple MSI-X */
- BNX2X_DEV_INFO("Failed to enable multiple MSI-X (%d), set number of queues to %d\n",
- bp->num_queues,
- 1 + bp->num_cnic_queues);
-
- bp->num_queues = 1 + bp->num_cnic_queues;
-
- /* Try to enable MSI */
- if (!(bp->flags & USING_SINGLE_MSIX_FLAG) &&
- !(bp->flags & DISABLE_MSI_FLAG))
- bnx2x_enable_msi(bp);
- }
- break;
+ BNX2X_DEV_INFO("unknown value in int_mode module parameter\n");
+ return -EINVAL;
}
+ return 0;
}
-/* must be called prioir to any HW initializations */
+/* must be called prior to any HW initializations */
static inline u16 bnx2x_cid_ilt_lines(struct bnx2x *bp)
{
return L2_ILT_LINES(bp);
@@ -11081,9 +11101,13 @@ static int bnx2x_init_bp(struct bnx2x *bp)
INIT_DELAYED_WORK(&bp->sp_task, bnx2x_sp_task);
INIT_DELAYED_WORK(&bp->sp_rtnl_task, bnx2x_sp_rtnl_task);
INIT_DELAYED_WORK(&bp->period_task, bnx2x_period_task);
- rc = bnx2x_get_hwinfo(bp);
- if (rc)
- return rc;
+ if (IS_PF(bp)) {
+ rc = bnx2x_get_hwinfo(bp);
+ if (rc)
+ return rc;
+ } else {
+ random_ether_addr(bp->dev->dev_addr);
+ }
bnx2x_set_modes_bitmap(bp);
@@ -11096,7 +11120,7 @@ static int bnx2x_init_bp(struct bnx2x *bp)
func = BP_FUNC(bp);
/* need to reset chip if undi was active */
- if (!BP_NOMCP(bp)) {
+ if (IS_PF(bp) && !BP_NOMCP(bp)) {
/* init fw_seq */
bp->fw_seq =
SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_mb_header) &
@@ -11133,6 +11157,8 @@ static int bnx2x_init_bp(struct bnx2x *bp)
bp->mrrs = mrrs;
bp->tx_ring_size = IS_MF_FCOE_AFEX(bp) ? 0 : MAX_TX_AVAIL;
+ if (IS_VF(bp))
+ bp->rx_ring_size = MAX_RX_AVAIL;
/* make sure that the numbers are in the right granularity */
bp->tx_ticks = (50 / BNX2X_BTR) * BNX2X_BTR;
@@ -11161,12 +11187,18 @@ static int bnx2x_init_bp(struct bnx2x *bp)
bp->cnic_base_cl_id = FP_SB_MAX_E2;
/* multiple tx priority */
- if (CHIP_IS_E1x(bp))
+ if (IS_VF(bp))
+ bp->max_cos = 1;
+ else if (CHIP_IS_E1x(bp))
bp->max_cos = BNX2X_MULTI_TX_COS_E1X;
- if (CHIP_IS_E2(bp) || CHIP_IS_E3A0(bp))
+ else if (CHIP_IS_E2(bp) || CHIP_IS_E3A0(bp))
bp->max_cos = BNX2X_MULTI_TX_COS_E2_E3A0;
- if (CHIP_IS_E3B0(bp))
+ else if (CHIP_IS_E3B0(bp))
bp->max_cos = BNX2X_MULTI_TX_COS_E3B0;
+ else
+ BNX2X_ERR("unknown chip %x revision %x\n",
+ CHIP_NUM(bp), CHIP_REV(bp));
+ BNX2X_DEV_INFO("set bp->max_cos to %d\n", bp->max_cos);
/* We need at least one default status block for slow-path events,
* second status block for the L2 queue, and a third status block for
@@ -11551,10 +11583,9 @@ static int bnx2x_set_coherency_mask(struct bnx2x *bp)
return 0;
}
-static int bnx2x_init_dev(struct pci_dev *pdev, struct net_device *dev,
- unsigned long board_type)
+static int bnx2x_init_dev(struct bnx2x *bp, struct pci_dev *pdev,
+ struct net_device *dev, unsigned long board_type)
{
- struct bnx2x *bp;
int rc;
u32 pci_cfg_dword;
bool chip_is_e1x = (board_type == BCM57710 ||
@@ -11562,11 +11593,9 @@ static int bnx2x_init_dev(struct pci_dev *pdev, struct net_device *dev,
board_type == BCM57711E);
SET_NETDEV_DEV(dev, &pdev->dev);
- bp = netdev_priv(dev);
bp->dev = dev;
bp->pdev = pdev;
- bp->flags = 0;
rc = pci_enable_device(pdev);
if (rc) {
@@ -11582,9 +11611,8 @@ static int bnx2x_init_dev(struct pci_dev *pdev, struct net_device *dev,
goto err_out_disable;
}
- if (!(pci_resource_flags(pdev, 2) & IORESOURCE_MEM)) {
- dev_err(&bp->pdev->dev, "Cannot find second PCI device"
- " base address, aborting\n");
+ if (IS_PF(bp) && !(pci_resource_flags(pdev, 2) & IORESOURCE_MEM)) {
+ dev_err(&bp->pdev->dev, "Cannot find second PCI device base address, aborting\n");
rc = -ENODEV;
goto err_out_disable;
}
@@ -11609,12 +11637,14 @@ static int bnx2x_init_dev(struct pci_dev *pdev, struct net_device *dev,
pci_save_state(pdev);
}
- bp->pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM);
- if (bp->pm_cap == 0) {
- dev_err(&bp->pdev->dev,
- "Cannot find power management capability, aborting\n");
- rc = -EIO;
- goto err_out_release;
+ if (IS_PF(bp)) {
+ bp->pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM);
+ if (bp->pm_cap == 0) {
+ dev_err(&bp->pdev->dev,
+ "Cannot find power management capability, aborting\n");
+ rc = -EIO;
+ goto err_out_release;
+ }
}
if (!pci_is_pcie(pdev)) {
@@ -11665,25 +11695,28 @@ static int bnx2x_init_dev(struct pci_dev *pdev, struct net_device *dev,
* Clean the following indirect addresses for all functions since it
* is not used by the driver.
*/
- REG_WR(bp, PXP2_REG_PGL_ADDR_88_F0, 0);
- REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F0, 0);
- REG_WR(bp, PXP2_REG_PGL_ADDR_90_F0, 0);
- REG_WR(bp, PXP2_REG_PGL_ADDR_94_F0, 0);
+ if (IS_PF(bp)) {
+ REG_WR(bp, PXP2_REG_PGL_ADDR_88_F0, 0);
+ REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F0, 0);
+ REG_WR(bp, PXP2_REG_PGL_ADDR_90_F0, 0);
+ REG_WR(bp, PXP2_REG_PGL_ADDR_94_F0, 0);
+
+ if (chip_is_e1x) {
+ REG_WR(bp, PXP2_REG_PGL_ADDR_88_F1, 0);
+ REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F1, 0);
+ REG_WR(bp, PXP2_REG_PGL_ADDR_90_F1, 0);
+ REG_WR(bp, PXP2_REG_PGL_ADDR_94_F1, 0);
+ }
- if (chip_is_e1x) {
- REG_WR(bp, PXP2_REG_PGL_ADDR_88_F1, 0);
- REG_WR(bp, PXP2_REG_PGL_ADDR_8C_F1, 0);
- REG_WR(bp, PXP2_REG_PGL_ADDR_90_F1, 0);
- REG_WR(bp, PXP2_REG_PGL_ADDR_94_F1, 0);
+ /* Enable internal target-read (in case we are probed after PF
+ * FLR). Must be done prior to any BAR read access. Only for
+ * 57712 and up
+ */
+ if (!chip_is_e1x)
+ REG_WR(bp,
+ PGLUE_B_REG_INTERNAL_PFID_ENABLE_TARGET_READ, 1);
}
- /*
- * Enable internal target-read (in case we are probed after PF FLR).
- * Must be done prior to any BAR read access. Only for 57712 and up
- */
- if (!chip_is_e1x)
- REG_WR(bp, PGLUE_B_REG_INTERNAL_PFID_ENABLE_TARGET_READ, 1);
-
dev->watchdog_timeo = TX_TIMEOUT;
dev->netdev_ops = &bnx2x_netdev_ops;
@@ -11734,8 +11767,9 @@ err_out:
static void bnx2x_get_pcie_width_speed(struct bnx2x *bp, int *width, int *speed)
{
- u32 val = REG_RD(bp, PCICFG_OFFSET + PCICFG_LINK_CONTROL);
+ u32 val = 0;
+ pci_read_config_dword(bp->pdev, PCICFG_LINK_CONTROL, &val);
*width = (val & PCICFG_LINK_WIDTH) >> PCICFG_LINK_WIDTH_SHIFT;
/* return value of 1=2.5GHz 2=5GHz */
@@ -12012,10 +12046,10 @@ static int bnx2x_set_qm_cid_count(struct bnx2x *bp)
*
*/
static int bnx2x_get_num_non_def_sbs(struct pci_dev *pdev,
- int cnic_cnt)
+ int cnic_cnt, bool is_vf)
{
- int pos;
- u16 control;
+ int pos, index;
+ u16 control = 0;
pos = pci_find_capability(pdev, PCI_CAP_ID_MSIX);
@@ -12023,85 +12057,114 @@ static int bnx2x_get_num_non_def_sbs(struct pci_dev *pdev,
* If MSI-X is not supported - return number of SBs needed to support
* one fast path queue: one FP queue + SB for CNIC
*/
- if (!pos)
+ if (!pos) {
+ dev_info(&pdev->dev, "no msix capability found\n");
return 1 + cnic_cnt;
+ }
+ dev_info(&pdev->dev, "msix capability found\n");
/*
* The value in the PCI configuration space is the index of the last
* entry, namely one less than the actual size of the table, which is
* exactly what we want to return from this function: number of all SBs
* without the default SB.
+ * For VFs there is no default SB, then we return (index+1).
*/
pci_read_config_word(pdev, pos + PCI_MSI_FLAGS, &control);
- return control & PCI_MSIX_FLAGS_QSIZE;
-}
-struct cnic_eth_dev *bnx2x_cnic_probe(struct net_device *);
+ index = control & PCI_MSIX_FLAGS_QSIZE;
-static int bnx2x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
-{
- struct net_device *dev = NULL;
- struct bnx2x *bp;
- int pcie_width, pcie_speed;
- int rc, max_non_def_sbs;
- int rx_count, tx_count, rss_count, doorbell_size;
- int cnic_cnt;
- /*
- * An estimated maximum supported CoS number according to the chip
- * version.
- * We will try to roughly estimate the maximum number of CoSes this chip
- * may support in order to minimize the memory allocated for Tx
- * netdev_queue's. This number will be accurately calculated during the
- * initialization of bp->max_cos based on the chip versions AND chip
- * revision in the bnx2x_init_bp().
- */
- u8 max_cos_est = 0;
+ return is_vf ? index + 1 : index;
+}
- switch (ent->driver_data) {
+static int set_max_cos_est(int chip_id)
+{
+ switch (chip_id) {
case BCM57710:
case BCM57711:
case BCM57711E:
- max_cos_est = BNX2X_MULTI_TX_COS_E1X;
- break;
-
+ return BNX2X_MULTI_TX_COS_E1X;
case BCM57712:
case BCM57712_MF:
- max_cos_est = BNX2X_MULTI_TX_COS_E2_E3A0;
- break;
-
+ case BCM57712_VF:
+ return BNX2X_MULTI_TX_COS_E2_E3A0;
case BCM57800:
case BCM57800_MF:
+ case BCM57800_VF:
case BCM57810:
case BCM57810_MF:
- case BCM57840_O:
case BCM57840_4_10:
case BCM57840_2_20:
+ case BCM57840_O:
case BCM57840_MFO:
+ case BCM57810_VF:
case BCM57840_MF:
+ case BCM57840_VF:
case BCM57811:
case BCM57811_MF:
- max_cos_est = BNX2X_MULTI_TX_COS_E3B0;
- break;
-
+ case BCM57811_VF:
+ return BNX2X_MULTI_TX_COS_E3B0;
+ return 1;
default:
- pr_err("Unknown board_type (%ld), aborting\n",
- ent->driver_data);
+ pr_err("Unknown board_type (%d), aborting\n", chip_id);
return -ENODEV;
}
+}
- cnic_cnt = 1;
- max_non_def_sbs = bnx2x_get_num_non_def_sbs(pdev, cnic_cnt);
+static int set_is_vf(int chip_id)
+{
+ switch (chip_id) {
+ case BCM57712_VF:
+ case BCM57800_VF:
+ case BCM57810_VF:
+ case BCM57840_VF:
+ case BCM57811_VF:
+ return true;
+ default:
+ return false;
+ }
+}
- WARN_ON(!max_non_def_sbs);
+struct cnic_eth_dev *bnx2x_cnic_probe(struct net_device *dev);
+
+static int bnx2x_init_one(struct pci_dev *pdev,
+ const struct pci_device_id *ent)
+{
+ struct net_device *dev = NULL;
+ struct bnx2x *bp;
+ int pcie_width, pcie_speed;
+ int rc, max_non_def_sbs;
+ int rx_count, tx_count, rss_count, doorbell_size;
+ int max_cos_est;
+ bool is_vf;
+ int cnic_cnt;
+
+ /* An estimated maximum supported CoS number according to the chip
+ * version.
+ * We will try to roughly estimate the maximum number of CoSes this chip
+ * may support in order to minimize the memory allocated for Tx
+ * netdev_queue's. This number will be accurately calculated during the
+ * initialization of bp->max_cos based on the chip versions AND chip
+ * revision in the bnx2x_init_bp().
+ */
+ max_cos_est = set_max_cos_est(ent->driver_data);
+ if (max_cos_est < 0)
+ return max_cos_est;
+ is_vf = set_is_vf(ent->driver_data);
+ cnic_cnt = is_vf ? 0 : 1;
+
+ max_non_def_sbs = bnx2x_get_num_non_def_sbs(pdev, cnic_cnt, is_vf);
/* Maximum number of RSS queues: one IGU SB goes to CNIC */
- rss_count = max_non_def_sbs - cnic_cnt;
+ rss_count = is_vf ? 1 : max_non_def_sbs - cnic_cnt;
+
+ if (rss_count < 1)
+ return -EINVAL;
/* Maximum number of netdev Rx queues: RSS + FCoE L2 */
rx_count = rss_count + cnic_cnt;
- /*
- * Maximum number of netdev Tx queues:
+ /* Maximum number of netdev Tx queues:
* Maximum TSS queues * Maximum supported number of CoS + FCoE L2
*/
tx_count = rss_count * max_cos_est + cnic_cnt;
@@ -12113,22 +12176,28 @@ static int bnx2x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
bp = netdev_priv(dev);
+ bp->flags = 0;
+ if (is_vf)
+ bp->flags |= IS_VF_FLAG;
+
bp->igu_sb_cnt = max_non_def_sbs;
+ bp->igu_base_addr = IS_VF(bp) ? PXP_VF_ADDR_IGU_START : BAR_IGU_INTMEM;
bp->msg_enable = debug;
bp->cnic_support = cnic_cnt;
bp->cnic_probe = bnx2x_cnic_probe;
pci_set_drvdata(pdev, dev);
- rc = bnx2x_init_dev(pdev, dev, ent->driver_data);
+ rc = bnx2x_init_dev(bp, pdev, dev, ent->driver_data);
if (rc < 0) {
free_netdev(dev);
return rc;
}
+ BNX2X_DEV_INFO("This is a %s function\n",
+ IS_PF(bp) ? "physical" : "virtual");
BNX2X_DEV_INFO("Cnic support is %s\n", CNIC_SUPPORT(bp) ? "on" : "off");
- BNX2X_DEV_INFO("max_non_def_sbs %d\n", max_non_def_sbs);
-
+ BNX2X_DEV_INFO("Max num of status blocks %d\n", max_non_def_sbs);
BNX2X_DEV_INFO("Allocated netdev with %d tx and %d rx queues\n",
tx_count, rx_count);
@@ -12136,19 +12205,28 @@ static int bnx2x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
if (rc)
goto init_one_exit;
- /*
- * Map doorbels here as we need the real value of bp->max_cos which
- * is initialized in bnx2x_init_bp().
+ /* Map doorbells here as we need the real value of bp->max_cos which
+ * is initialized in bnx2x_init_bp() to determine the number of
+ * l2 connections.
*/
- doorbell_size = BNX2X_L2_MAX_CID(bp) * (1 << BNX2X_DB_SHIFT);
- if (doorbell_size > pci_resource_len(pdev, 2)) {
- dev_err(&bp->pdev->dev,
- "Cannot map doorbells, bar size too small, aborting\n");
- rc = -ENOMEM;
- goto init_one_exit;
+ if (IS_VF(bp)) {
+ /* vf doorbells are embedded within the regview */
+ bp->doorbells = bp->regview + PXP_VF_ADDR_DB_START;
+
+ /* allocate vf2pf mailbox for vf to pf channel */
+ BNX2X_PCI_ALLOC(bp->vf2pf_mbox, &bp->vf2pf_mbox_mapping,
+ sizeof(struct bnx2x_vf_mbx_msg));
+ } else {
+ doorbell_size = BNX2X_L2_MAX_CID(bp) * (1 << BNX2X_DB_SHIFT);
+ if (doorbell_size > pci_resource_len(pdev, 2)) {
+ dev_err(&bp->pdev->dev,
+ "Cannot map doorbells, bar size too small, aborting\n");
+ rc = -ENOMEM;
+ goto init_one_exit;
+ }
+ bp->doorbells = ioremap_nocache(pci_resource_start(pdev, 2),
+ doorbell_size);
}
- bp->doorbells = ioremap_nocache(pci_resource_start(pdev, 2),
- doorbell_size);
if (!bp->doorbells) {
dev_err(&bp->pdev->dev,
"Cannot map doorbell space, aborting\n");
@@ -12158,6 +12236,7 @@ static int bnx2x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
/* calc qm_cid_count */
bp->qm_cid_count = bnx2x_set_qm_cid_count(bp);
+ BNX2X_DEV_INFO("qm_cid_count %d\n", bp->qm_cid_count);
/* disable FCOE L2 queue for E1x*/
if (CHIP_IS_E1x(bp))
@@ -12179,13 +12258,19 @@ static int bnx2x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
/* Configure interrupt mode: try to enable MSI-X/MSI if
* needed.
*/
- bnx2x_set_int_mode(bp);
+ rc = bnx2x_set_int_mode(bp);
+ if (rc) {
+ dev_err(&pdev->dev, "Cannot set interrupts\n");
+ goto init_one_exit;
+ }
+ /* register the net device */
rc = register_netdev(dev);
if (rc) {
dev_err(&pdev->dev, "Cannot register net device\n");
goto init_one_exit;
}
+ BNX2X_DEV_INFO("device name after netdev register %s\n", dev->name);
if (!NO_FCOE(bp)) {
@@ -12196,6 +12281,8 @@ static int bnx2x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
}
bnx2x_get_pcie_width_speed(bp, &pcie_width, &pcie_speed);
+ BNX2X_DEV_INFO("got pcie width %d and speed %d\n",
+ pcie_width, pcie_speed);
BNX2X_DEV_INFO(
"%s (%c%d) PCI-E x%d %s found at mem %lx, IRQ %d, node addr %pM\n",
@@ -12209,11 +12296,16 @@ static int bnx2x_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
return 0;
+alloc_mem_err:
+ BNX2X_PCI_FREE(bp->vf2pf_mbox, bp->vf2pf_mbox_mapping,
+ sizeof(struct bnx2x_vf_mbx_msg));
+ rc = -ENOMEM;
+
init_one_exit:
if (bp->regview)
iounmap(bp->regview);
- if (bp->doorbells)
+ if (IS_PF(bp) && bp->doorbells)
iounmap(bp->doorbells);
free_netdev(dev);
@@ -12253,13 +12345,15 @@ static void bnx2x_remove_one(struct pci_dev *pdev)
unregister_netdev(dev);
/* Power on: we can't let PCI layer write to us while we are in D3 */
- bnx2x_set_power_state(bp, PCI_D0);
+ if (IS_PF(bp))
+ bnx2x_set_power_state(bp, PCI_D0);
/* Disable MSI/MSI-X */
bnx2x_disable_msi(bp);
/* Power off */
- bnx2x_set_power_state(bp, PCI_D3hot);
+ if (IS_PF(bp))
+ bnx2x_set_power_state(bp, PCI_D3hot);
/* Make sure RESET task is not scheduled before continuing */
cancel_delayed_work_sync(&bp->sp_rtnl_task);
@@ -12267,11 +12361,15 @@ static void bnx2x_remove_one(struct pci_dev *pdev)
if (bp->regview)
iounmap(bp->regview);
- if (bp->doorbells)
- iounmap(bp->doorbells);
-
- bnx2x_release_firmware(bp);
+ /* for vf doorbells are part of the regview and were unmapped along with
+ * it. FW is only loaded by PF.
+ */
+ if (IS_PF(bp)) {
+ if (bp->doorbells)
+ iounmap(bp->doorbells);
+ bnx2x_release_firmware(bp);
+ }
bnx2x_free_mem_bp(bp);
free_netdev(dev);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index bc2f65b..463a984 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -6554,6 +6554,15 @@
(7L<<ME_REG_ABS_PF_NUM_SHIFT) /* Absolute PF Num */
+#define PXP_VF_ADDR_IGU_START 0
+#define PXP_VF_ADDR_IGU_SIZE 0x3000
+#define PXP_VF_ADDR_IGU_END\
+ ((PXP_VF_ADDR_IGU_START) + (PXP_VF_ADDR_IGU_SIZE) - 1)
+#define PXP_VF_ADDR_DB_START 0x7c00
+#define PXP_VF_ADDR_DB_SIZE 0x200
+#define PXP_VF_ADDR_DB_END\
+ ((PXP_VF_ADDR_DB_START) + (PXP_VF_ADDR_DB_SIZE) - 1)
+
#define MDIO_REG_BANK_CL73_IEEEB0 0x0
#define MDIO_CL73_IEEEB0_CL73_AN_CONTROL 0x0
#define MDIO_CL73_IEEEB0_CL73_AN_CONTROL_RESTART_AN 0x0200
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
new file mode 100644
index 0000000..1b14745
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.h
@@ -0,0 +1,27 @@
+/* bnx2x_sriov.h: Broadcom Everest network driver.
+ *
+ * Copyright 2009-2012 Broadcom Corporation
+ *
+ * Unless you and Broadcom execute a separate written software license
+ * agreement governing use of this software, this software is licensed to you
+ * under the terms of the GNU General Public License version 2, available
+ * at http://www.gnu.org/licenses/old-licenses/gpl-2.0.html (the "GPL").
+ *
+ * Notwithstanding the above, under no circumstances may you combine this
+ * software in any way with any other Broadcom software provided under a
+ * license other than the GPL, without Broadcom's express prior written
+ * consent.
+ *
+ * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Written by: Shmulik Ravid <shmulikr@broadcom.com>
+ * Ariel Elior <ariele@broadcom.com>
+ */
+#ifndef BNX2X_SRIOV_H
+#define BNX2X_SRIOV_H
+
+struct bnx2x_vf_mbx_msg {
+ union vfpf_tlvs req;
+ union pfvf_tlvs resp;
+};
+
+#endif /* bnx2x_sriov.h */
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
new file mode 100644
index 0000000..bb37675
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
@@ -0,0 +1,37 @@
+/* bnx2x_vfpf.h: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2011-2012 Broadcom Corporation
+ *
+ * Unless you and Broadcom execute a separate written software license
+ * agreement governing use of this software, this software is licensed to you
+ * under the terms of the GNU General Public License version 2, available
+ * at http://www.gnu.org/licenses/old-licenses/gpl-2.0.html (the "GPL").
+ *
+ * Notwithstanding the above, under no circumstances may you combine this
+ * software in any way with any other Broadcom software provided under a
+ * license other than the GPL, without Broadcom's express prior written
+ * consent.
+ *
+ * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Written by: Ariel Elior <ariele@broadcom.com>
+ */
+#ifndef VF_PF_IF_H
+#define VF_PF_IF_H
+
+/* HW VF-PF channel definitions
+ * A.K.A VF-PF mailbox
+ */
+#define TLV_BUFFER_SIZE 1024
+
+struct tlv_buffer_size {
+ u8 tlv_buffer[TLV_BUFFER_SIZE];
+};
+
+union vfpf_tlvs {
+ struct tlv_buffer_size tlv_buf_size;
+};
+
+union pfvf_tlvs {
+ struct tlv_buffer_size tlv_buf_size;
+};
+#endif /* VF_PF_IF_H */
--
1.7.9.GIT
^ permalink raw reply related
* [PATCH net-next v4 05/22] bnx2x: Add init, setup_q, set_mac to VF <-> PF channel
From: Ariel Elior @ 2012-12-30 19:57 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Ariel Elior, Eilon Greenstein
In-Reply-To: <1356897442-3063-1-git-send-email-ariele@broadcom.com>
'init' - init an acquired VF. Supply allocation GPAs to PF.
'setup_q' - PF to allocate a queue in device on behalf of the VF.
'set_mac' - PF to configure a mac in device on behalf of the VF.
VF driver uses these requests in the VF <-> PF channel in nic_load
flow.
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 6 +
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 18 +++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 163 ++++++++++++++++++++++
drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h | 123 ++++++++++++++++
4 files changed, 310 insertions(+), 0 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index 634eb30..5c64d1d 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -341,6 +341,9 @@ union db_prod {
#define SGE_PAGE_SIZE PAGE_SIZE
#define SGE_PAGE_SHIFT PAGE_SHIFT
#define SGE_PAGE_ALIGN(addr) PAGE_ALIGN((typeof(PAGE_SIZE))(addr))
+#define SGE_PAGES (SGE_PAGE_SIZE * PAGES_PER_SGE)
+#define TPA_AGG_SIZE min_t(u32, (min_t(u32, 8, MAX_SKB_FRAGS) * \
+ SGE_PAGES), 0xffff)
/* SGE ring related macros */
#define NUM_RX_SGE_PAGES 2
@@ -2221,6 +2224,9 @@ int bnx2x_get_vf_id(struct bnx2x *bp, u32 *vf_id);
int bnx2x_send_msg2pf(struct bnx2x *bp, u8 *done, dma_addr_t msg_mapping);
int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count);
int bnx2x_vfpf_release(struct bnx2x *bp);
+int bnx2x_vfpf_init(struct bnx2x *bp);
+int bnx2x_vfpf_setup_q(struct bnx2x *bp, int fp_idx);
+int bnx2x_vfpf_set_mac(struct bnx2x *bp);
int bnx2x_nic_load_analyze_req(struct bnx2x *bp, u32 load_code);
/* Congestion management fairness mode */
#define CMNG_FNS_NONE 0
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 646bfd1..f7b23c2 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -2445,6 +2445,13 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
LOAD_ERROR_EXIT(bp, load_error0);
}
+ /* request pf to initialize status blocks */
+ if (IS_VF(bp)) {
+ rc = bnx2x_vfpf_init(bp);
+ if (rc)
+ LOAD_ERROR_EXIT(bp, load_error0);
+ }
+
/* As long as bnx2x_alloc_mem() may possibly update
* bp->num_queues, bnx2x_set_real_num_queues() should always
* come after it. At this stage cnic queues are not counted.
@@ -2564,6 +2571,15 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
BNX2X_ERR("PF RSS init failed\n");
LOAD_ERROR_EXIT(bp, load_error3);
}
+
+ } else { /* vf */
+ for_each_eth_queue(bp, i) {
+ rc = bnx2x_vfpf_setup_q(bp, i);
+ if (rc) {
+ BNX2X_ERR("Queue setup failed\n");
+ LOAD_ERROR_EXIT(bp, load_error3);
+ }
+ }
}
/* Now when Clients are configured we are ready to work */
@@ -2572,6 +2588,8 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
/* Configure a ucast MAC */
if (IS_PF(bp))
rc = bnx2x_set_eth_mac(bp, true);
+ else /* vf */
+ rc = bnx2x_vfpf_set_mac(bp);
if (rc) {
BNX2X_ERR("Setting Ethernet MAC failed\n");
LOAD_ERROR_EXIT(bp, load_error3);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 97b8aa7..9a41072 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -13431,3 +13431,166 @@ int bnx2x_vfpf_release(struct bnx2x *bp)
return 0;
}
+
+/* Tell PF about SB addresses */
+int bnx2x_vfpf_init(struct bnx2x *bp)
+{
+ struct vfpf_init_tlv *req = &bp->vf2pf_mbox->req.init;
+ struct pfvf_general_resp_tlv *resp = &bp->vf2pf_mbox->resp.general_resp;
+ int rc, i;
+
+ /* clear mailbox and prep first tlv */
+ bnx2x_vfpf_prep(bp, &req->first_tlv, CHANNEL_TLV_INIT, sizeof(*req));
+
+ /* status blocks */
+ for_each_eth_queue(bp, i)
+ req->sb_addr[i] = (dma_addr_t)bnx2x_fp(bp, i,
+ status_blk_mapping);
+
+ /* statistics - requests only supports single queue for now */
+ req->stats_addr = bp->fw_stats_data_mapping +
+ offsetof(struct bnx2x_fw_stats_data, queue_stats);
+
+ /* add list termination tlv */
+ bnx2x_add_tlv(bp, req, req->first_tlv.tl.length, CHANNEL_TLV_LIST_END,
+ sizeof(struct channel_list_end_tlv));
+
+ /* output tlvs list */
+ bnx2x_dp_tlv_list(bp, req);
+
+ rc = bnx2x_send_msg2pf(bp, &resp->hdr.status, bp->vf2pf_mbox_mapping);
+ if (rc)
+ return rc;
+
+ if (resp->hdr.status != PFVF_STATUS_SUCCESS) {
+ BNX2X_ERR("INIT VF failed: %d. Breaking...\n",
+ resp->hdr.status);
+ return -EAGAIN;
+ }
+
+ DP(BNX2X_MSG_SP, "INIT VF Succeeded\n");
+ return 0;
+}
+
+/* ask the pf to open a queue for the vf */
+int bnx2x_vfpf_setup_q(struct bnx2x *bp, int fp_idx)
+{
+ struct vfpf_setup_q_tlv *req = &bp->vf2pf_mbox->req.setup_q;
+ struct pfvf_general_resp_tlv *resp = &bp->vf2pf_mbox->resp.general_resp;
+ struct bnx2x_fastpath *fp = &bp->fp[fp_idx];
+ u16 tpa_agg_size = 0, flags = 0;
+ int rc;
+
+ /* clear mailbox and prep first tlv */
+ bnx2x_vfpf_prep(bp, &req->first_tlv, CHANNEL_TLV_SETUP_Q, sizeof(*req));
+
+ /* select tpa mode to request */
+ if (!fp->disable_tpa) {
+ flags |= VFPF_QUEUE_FLG_TPA;
+ flags |= VFPF_QUEUE_FLG_TPA_IPV6;
+ if (fp->mode == TPA_MODE_GRO)
+ flags |= VFPF_QUEUE_FLG_TPA_GRO;
+ tpa_agg_size = TPA_AGG_SIZE;
+ }
+
+ /* calculate queue flags */
+ flags |= VFPF_QUEUE_FLG_STATS;
+ flags |= VFPF_QUEUE_FLG_CACHE_ALIGN;
+ flags |= IS_MF_SD(bp) ? VFPF_QUEUE_FLG_OV : 0;
+ flags |= VFPF_QUEUE_FLG_VLAN;
+ DP(NETIF_MSG_IFUP, "vlan removal enabled\n");
+
+ /* Common */
+ req->vf_qid = fp_idx;
+ req->param_valid = VFPF_RXQ_VALID | VFPF_TXQ_VALID;
+
+ /* Rx */
+ req->rxq.rcq_addr = fp->rx_comp_mapping;
+ req->rxq.rcq_np_addr = fp->rx_comp_mapping + BCM_PAGE_SIZE;
+ req->rxq.rxq_addr = fp->rx_desc_mapping;
+ req->rxq.sge_addr = fp->rx_sge_mapping;
+ req->rxq.vf_sb = fp_idx;
+ req->rxq.sb_index = HC_INDEX_ETH_RX_CQ_CONS;
+ req->rxq.hc_rate = bp->rx_ticks ? 1000000/bp->rx_ticks : 0;
+ req->rxq.mtu = bp->dev->mtu;
+ req->rxq.buf_sz = fp->rx_buf_size;
+ req->rxq.sge_buf_sz = BCM_PAGE_SIZE * PAGES_PER_SGE;
+ req->rxq.tpa_agg_sz = tpa_agg_size;
+ req->rxq.max_sge_pkt = SGE_PAGE_ALIGN(bp->dev->mtu) >> SGE_PAGE_SHIFT;
+ req->rxq.max_sge_pkt = ((req->rxq.max_sge_pkt + PAGES_PER_SGE - 1) &
+ (~(PAGES_PER_SGE-1))) >> PAGES_PER_SGE_SHIFT;
+ req->rxq.flags = flags;
+ req->rxq.drop_flags = 0;
+ req->rxq.cache_line_log = BNX2X_RX_ALIGN_SHIFT;
+ req->rxq.stat_id = -1; /* No stats at the moment */
+
+ /* Tx */
+ req->txq.txq_addr = fp->txdata_ptr[FIRST_TX_COS_INDEX]->tx_desc_mapping;
+ req->txq.vf_sb = fp_idx;
+ req->txq.sb_index = HC_INDEX_ETH_TX_CQ_CONS_COS0;
+ req->txq.hc_rate = bp->tx_ticks ? 1000000/bp->tx_ticks : 0;
+ req->txq.flags = flags;
+ req->txq.traffic_type = LLFC_TRAFFIC_TYPE_NW;
+
+ /* add list termination tlv */
+ bnx2x_add_tlv(bp, req, req->first_tlv.tl.length, CHANNEL_TLV_LIST_END,
+ sizeof(struct channel_list_end_tlv));
+
+ /* output tlvs list */
+ bnx2x_dp_tlv_list(bp, req);
+
+ rc = bnx2x_send_msg2pf(bp, &resp->hdr.status, bp->vf2pf_mbox_mapping);
+ if (rc)
+ BNX2X_ERR("Sending SETUP_Q message for queue[%d] failed!\n",
+ fp_idx);
+
+ if (resp->hdr.status != PFVF_STATUS_SUCCESS) {
+ BNX2X_ERR("Status of SETUP_Q for queue[%d] is %d\n",
+ fp_idx, resp->hdr.status);
+ return -EINVAL;
+ }
+ return rc;
+}
+
+/* request pf to add a mac for the vf */
+int bnx2x_vfpf_set_mac(struct bnx2x *bp)
+{
+ struct vfpf_set_q_filters_tlv *req = &bp->vf2pf_mbox->req.set_q_filters;
+ struct pfvf_general_resp_tlv *resp = &bp->vf2pf_mbox->resp.general_resp;
+ int rc;
+
+ /* clear mailbox and prep first tlv */
+ bnx2x_vfpf_prep(bp, &req->first_tlv, CHANNEL_TLV_SET_Q_FILTERS,
+ sizeof(*req));
+
+ req->flags = VFPF_SET_Q_FILTERS_MAC_VLAN_CHANGED;
+ req->vf_qid = 0;
+ req->n_mac_vlan_filters = 1;
+ req->filters[0].flags =
+ VFPF_Q_FILTER_DEST_MAC_VALID | VFPF_Q_FILTER_SET_MAC;
+
+ /* copy mac from device to request */
+ memcpy(req->filters[0].mac, bp->dev->dev_addr, ETH_ALEN);
+
+ /* add list termination tlv */
+ bnx2x_add_tlv(bp, req, req->first_tlv.tl.length, CHANNEL_TLV_LIST_END,
+ sizeof(struct channel_list_end_tlv));
+
+ /* output tlvs list */
+ bnx2x_dp_tlv_list(bp, req);
+
+ /* send message to pf */
+ rc = bnx2x_send_msg2pf(bp, &resp->hdr.status, bp->vf2pf_mbox_mapping);
+ if (rc) {
+ BNX2X_ERR("failed to send message to pf. rc was %d\n", rc);
+ return rc;
+ }
+
+ /* PF failed the transaction */
+ if (resp->hdr.status != PFVF_STATUS_SUCCESS) {
+ BNX2X_ERR("vfpf SET MAC failed: %d\n", resp->hdr.status);
+ return -EINVAL;
+ }
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
index 561c656..12889f0 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.h
@@ -38,6 +38,22 @@ struct hw_sb_info {
*/
#define TLV_BUFFER_SIZE 1024
+#define VFPF_QUEUE_FLG_TPA 0x0001
+#define VFPF_QUEUE_FLG_TPA_IPV6 0x0002
+#define VFPF_QUEUE_FLG_TPA_GRO 0x0004
+#define VFPF_QUEUE_FLG_CACHE_ALIGN 0x0008
+#define VFPF_QUEUE_FLG_STATS 0x0010
+#define VFPF_QUEUE_FLG_OV 0x0020
+#define VFPF_QUEUE_FLG_VLAN 0x0040
+#define VFPF_QUEUE_FLG_COS 0x0080
+#define VFPF_QUEUE_FLG_HC 0x0100
+#define VFPF_QUEUE_FLG_DHC 0x0200
+
+#define VFPF_QUEUE_DROP_IP_CS_ERR (1 << 0)
+#define VFPF_QUEUE_DROP_TCP_CS_ERR (1 << 1)
+#define VFPF_QUEUE_DROP_TTL0 (1 << 2)
+#define VFPF_QUEUE_DROP_UDP_CS_ERR (1 << 3)
+
enum {
PFVF_STATUS_WAITING = 0,
PFVF_STATUS_SUCCESS,
@@ -130,6 +146,107 @@ struct pfvf_acquire_resp_tlv {
} resc;
};
+/* Init VF */
+struct vfpf_init_tlv {
+ struct vfpf_first_tlv first_tlv;
+ aligned_u64 sb_addr[PFVF_MAX_SBS_PER_VF]; /* vf_sb based */
+ aligned_u64 spq_addr;
+ aligned_u64 stats_addr;
+};
+
+/* Setup Queue */
+struct vfpf_setup_q_tlv {
+ struct vfpf_first_tlv first_tlv;
+
+ struct vf_pf_rxq_params {
+ /* physical addresses */
+ aligned_u64 rcq_addr;
+ aligned_u64 rcq_np_addr;
+ aligned_u64 rxq_addr;
+ aligned_u64 sge_addr;
+
+ /* sb + hc info */
+ u8 vf_sb; /* index in hw_sbs[] */
+ u8 sb_index; /* Index in the SB */
+ u16 hc_rate; /* desired interrupts per sec. */
+ /* valid iff VFPF_QUEUE_FLG_HC */
+ /* rx buffer info */
+ u16 mtu;
+ u16 buf_sz;
+ u16 flags; /* VFPF_QUEUE_FLG_X flags */
+ u16 stat_id; /* valid iff VFPF_QUEUE_FLG_STATS */
+
+ /* valid iff VFPF_QUEUE_FLG_TPA */
+ u16 sge_buf_sz;
+ u16 tpa_agg_sz;
+ u8 max_sge_pkt;
+
+ u8 drop_flags; /* VFPF_QUEUE_DROP_X, for Linux VMs
+ * all the flags are turned off
+ */
+
+ u8 cache_line_log; /* VFPF_QUEUE_FLG_CACHE_ALIGN */
+ u8 padding;
+ } rxq;
+
+ struct vf_pf_txq_params {
+ /* physical addresses */
+ aligned_u64 txq_addr;
+
+ /* sb + hc info */
+ u8 vf_sb; /* index in hw_sbs[] */
+ u8 sb_index; /* Index in the SB */
+ u16 hc_rate; /* desired interrupts per sec. */
+ /* valid iff VFPF_QUEUE_FLG_HC */
+ u32 flags; /* VFPF_QUEUE_FLG_X flags */
+ u16 stat_id; /* valid iff VFPF_QUEUE_FLG_STATS */
+ u8 traffic_type; /* see in setup_context() */
+ u8 padding;
+ } txq;
+
+ u8 vf_qid; /* index in hw_qid[] */
+ u8 param_valid;
+#define VFPF_RXQ_VALID 0x01
+#define VFPF_TXQ_VALID 0x02
+ u8 padding[2];
+};
+
+/* Set Queue Filters */
+struct vfpf_q_mac_vlan_filter {
+ u32 flags;
+#define VFPF_Q_FILTER_DEST_MAC_VALID 0x01
+#define VFPF_Q_FILTER_VLAN_TAG_VALID 0x02
+#define VFPF_Q_FILTER_SET_MAC 0x100 /* set/clear */
+ u8 mac[ETH_ALEN];
+ u16 vlan_tag;
+};
+
+/* configure queue filters */
+struct vfpf_set_q_filters_tlv {
+ struct vfpf_first_tlv first_tlv;
+
+ u32 flags;
+#define VFPF_SET_Q_FILTERS_MAC_VLAN_CHANGED 0x01
+#define VFPF_SET_Q_FILTERS_MULTICAST_CHANGED 0x02
+#define VFPF_SET_Q_FILTERS_RX_MASK_CHANGED 0x04
+
+ u8 vf_qid; /* index in hw_qid[] */
+ u8 n_mac_vlan_filters;
+ u8 n_multicast;
+ u8 padding;
+
+#define PFVF_MAX_MAC_FILTERS 16
+#define PFVF_MAX_VLAN_FILTERS 16
+#define PFVF_MAX_FILTERS (PFVF_MAX_MAC_FILTERS +\
+ PFVF_MAX_VLAN_FILTERS)
+ struct vfpf_q_mac_vlan_filter filters[PFVF_MAX_FILTERS];
+
+#define PFVF_MAX_MULTICAST_PER_VF 32
+ u8 multicast[PFVF_MAX_MULTICAST_PER_VF][ETH_ALEN];
+
+ u32 rx_mask; /* see mask constants at the top of the file */
+};
+
/* release the VF's acquired resources */
struct vfpf_release_tlv {
struct vfpf_first_tlv first_tlv;
@@ -144,6 +261,9 @@ struct tlv_buffer_size {
union vfpf_tlvs {
struct vfpf_first_tlv first_tlv;
struct vfpf_acquire_tlv acquire;
+ struct vfpf_init_tlv init;
+ struct vfpf_setup_q_tlv setup_q;
+ struct vfpf_set_q_filters_tlv set_q_filters;
struct vfpf_release_tlv release;
struct channel_list_end_tlv list_end;
struct tlv_buffer_size tlv_buf_size;
@@ -161,6 +281,9 @@ union pfvf_tlvs {
enum channel_tlvs {
CHANNEL_TLV_NONE,
CHANNEL_TLV_ACQUIRE,
+ CHANNEL_TLV_INIT,
+ CHANNEL_TLV_SETUP_Q,
+ CHANNEL_TLV_SET_Q_FILTERS,
CHANNEL_TLV_RELEASE,
CHANNEL_TLV_LIST_END,
CHANNEL_TLV_MAX
--
1.7.9.GIT
^ permalink raw reply related
* [PATCH net-next v4 04/22] bnx2x: Separate VF and PF logic
From: Ariel Elior @ 2012-12-30 19:57 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Ariel Elior, Eilon Greenstein
In-Reply-To: <1356897442-3063-1-git-send-email-ariele@broadcom.com>
Generally, the VF driver cannot access the chip, except by the
narrow window its BAR allows. Care had to be taken so the VF driver
will not reach code which accesses the chip elsewhere.
Refactor the nic_load flow into parts so it would be
easier to separate the VF-only logic from the PF-only logic.
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 1 +
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 611 +++++++++++++-------
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 15 +-
.../net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c | 2 +-
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 95 +++-
drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h | 6 +
6 files changed, 496 insertions(+), 234 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
index 4dc754b..634eb30 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h
@@ -2221,6 +2221,7 @@ int bnx2x_get_vf_id(struct bnx2x *bp, u32 *vf_id);
int bnx2x_send_msg2pf(struct bnx2x *bp, u8 *done, dma_addr_t msg_mapping);
int bnx2x_vfpf_acquire(struct bnx2x *bp, u8 tx_count, u8 rx_count);
int bnx2x_vfpf_release(struct bnx2x *bp);
+int bnx2x_nic_load_analyze_req(struct bnx2x *bp, u32 load_code);
/* Congestion management fairness mode */
#define CMNG_FNS_NONE 0
#define CMNG_FNS_MINMAX 1
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 8687eb3..646bfd1 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -1064,7 +1064,7 @@ void __bnx2x_link_report(struct bnx2x *bp)
struct bnx2x_link_report_data cur_data;
/* reread mf_cfg */
- if (!CHIP_IS_E1(bp))
+ if (IS_PF(bp) && !CHIP_IS_E1(bp))
bnx2x_read_mf_cfg(bp);
/* Read the current link report info */
@@ -1406,10 +1406,14 @@ static void bnx2x_free_msix_irqs(struct bnx2x *bp, int nvecs)
if (nvecs == offset)
return;
- free_irq(bp->msix_table[offset].vector, bp->dev);
- DP(NETIF_MSG_IFDOWN, "released sp irq (%d)\n",
- bp->msix_table[offset].vector);
- offset++;
+
+ /* VFs don't have a default SB */
+ if (IS_PF(bp)) {
+ free_irq(bp->msix_table[offset].vector, bp->dev);
+ DP(NETIF_MSG_IFDOWN, "released sp irq (%d)\n",
+ bp->msix_table[offset].vector);
+ offset++;
+ }
if (CNIC_SUPPORT(bp)) {
if (nvecs == offset)
@@ -1430,11 +1434,17 @@ static void bnx2x_free_msix_irqs(struct bnx2x *bp, int nvecs)
void bnx2x_free_irq(struct bnx2x *bp)
{
if (bp->flags & USING_MSIX_FLAG &&
- !(bp->flags & USING_SINGLE_MSIX_FLAG))
- bnx2x_free_msix_irqs(bp, BNX2X_NUM_ETH_QUEUES(bp) +
- CNIC_SUPPORT(bp) + 1);
- else
+ !(bp->flags & USING_SINGLE_MSIX_FLAG)) {
+ int nvecs = BNX2X_NUM_ETH_QUEUES(bp) + CNIC_SUPPORT(bp);
+
+ /* vfs don't have a default status block */
+ if (IS_PF(bp))
+ nvecs++;
+
+ bnx2x_free_msix_irqs(bp, nvecs);
+ } else {
free_irq(bp->dev->irq, bp->dev);
+ }
}
int bnx2x_enable_msix(struct bnx2x *bp)
@@ -1530,12 +1540,15 @@ static int bnx2x_req_msix_irqs(struct bnx2x *bp)
{
int i, rc, offset = 0;
- rc = request_irq(bp->msix_table[offset++].vector,
- bnx2x_msix_sp_int, 0,
- bp->dev->name, bp->dev);
- if (rc) {
- BNX2X_ERR("request sp irq failed\n");
- return -EBUSY;
+ /* no default status block for vf */
+ if (IS_PF(bp)) {
+ rc = request_irq(bp->msix_table[offset++].vector,
+ bnx2x_msix_sp_int, 0,
+ bp->dev->name, bp->dev);
+ if (rc) {
+ BNX2X_ERR("request sp irq failed\n");
+ return -EBUSY;
+ }
}
if (CNIC_SUPPORT(bp))
@@ -1559,12 +1572,20 @@ static int bnx2x_req_msix_irqs(struct bnx2x *bp)
}
i = BNX2X_NUM_ETH_QUEUES(bp);
- offset = 1 + CNIC_SUPPORT(bp);
- netdev_info(bp->dev, "using MSI-X IRQs: sp %d fp[%d] %d ... fp[%d] %d\n",
- bp->msix_table[0].vector,
- 0, bp->msix_table[offset].vector,
- i - 1, bp->msix_table[offset + i - 1].vector);
-
+ if (IS_PF(bp)) {
+ offset = 1 + CNIC_SUPPORT(bp);
+ netdev_info(bp->dev,
+ "using MSI-X IRQs: sp %d fp[%d] %d ... fp[%d] %d\n",
+ bp->msix_table[0].vector,
+ 0, bp->msix_table[offset].vector,
+ i - 1, bp->msix_table[offset + i - 1].vector);
+ } else {
+ offset = CNIC_SUPPORT(bp);
+ netdev_info(bp->dev,
+ "using MSI-X IRQs: fp[%d] %d ... fp[%d] %d\n",
+ 0, bp->msix_table[offset].vector,
+ i - 1, bp->msix_table[offset + i - 1].vector);
+ }
return 0;
}
@@ -1972,27 +1993,204 @@ static void bnx2x_squeeze_objects(struct bnx2x *bp)
} while (0)
#endif /*BNX2X_STOP_ON_ERROR*/
-bool bnx2x_test_firmware_version(struct bnx2x *bp, bool is_err)
+static void bnx2x_free_fw_stats_mem(struct bnx2x *bp)
+{
+ BNX2X_PCI_FREE(bp->fw_stats, bp->fw_stats_mapping,
+ bp->fw_stats_data_sz + bp->fw_stats_req_sz);
+ return;
+}
+
+static int bnx2x_alloc_fw_stats_mem(struct bnx2x *bp)
{
- /* build FW version dword */
- u32 my_fw = (BCM_5710_FW_MAJOR_VERSION) +
- (BCM_5710_FW_MINOR_VERSION << 8) +
- (BCM_5710_FW_REVISION_VERSION << 16) +
- (BCM_5710_FW_ENGINEERING_VERSION << 24);
+ int num_groups;
+ int is_fcoe_stats = NO_FCOE(bp) ? 0 : 1;
- /* read loaded FW from chip */
- u32 loaded_fw = REG_RD(bp, XSEM_REG_PRAM);
+ /* number of queues for statistics is number of eth queues + FCoE */
+ u8 num_queue_stats = BNX2X_NUM_ETH_QUEUES(bp) + is_fcoe_stats;
- DP(NETIF_MSG_IFUP, "loaded fw %x, my fw %x\n", loaded_fw, my_fw);
+ /* Total number of FW statistics requests =
+ * 1 for port stats + 1 for PF stats + potential 2 for FCoE (fcoe proper
+ * and fcoe l2 queue) stats + num of queues (which includes another 1
+ * for fcoe l2 queue if applicable)
+ */
+ bp->fw_stats_num = 2 + is_fcoe_stats + num_queue_stats;
- if (loaded_fw != my_fw) {
- if (is_err)
- BNX2X_ERR("bnx2x with FW %x was already loaded, which mismatches my %x FW. aborting\n",
+ /* Request is built from stats_query_header and an array of
+ * stats_query_cmd_group each of which contains
+ * STATS_QUERY_CMD_COUNT rules. The real number or requests is
+ * configured in the stats_query_header.
+ */
+ num_groups =
+ (((bp->fw_stats_num) / STATS_QUERY_CMD_COUNT) +
+ (((bp->fw_stats_num) % STATS_QUERY_CMD_COUNT) ?
+ 1 : 0));
+
+ DP(BNX2X_MSG_SP, "stats fw_stats_num %d, num_groups %d\n",
+ bp->fw_stats_num, num_groups);
+ bp->fw_stats_req_sz = sizeof(struct stats_query_header) +
+ num_groups * sizeof(struct stats_query_cmd_group);
+
+ /* Data for statistics requests + stats_counter
+ * stats_counter holds per-STORM counters that are incremented
+ * when STORM has finished with the current request.
+ * memory for FCoE offloaded statistics are counted anyway,
+ * even if they will not be sent.
+ * VF stats are not accounted for here as the data of VF stats is stored
+ * in memory allocated by the VF, not here.
+ */
+ bp->fw_stats_data_sz = sizeof(struct per_port_stats) +
+ sizeof(struct per_pf_stats) +
+ sizeof(struct fcoe_statistics_params) +
+ sizeof(struct per_queue_stats) * num_queue_stats +
+ sizeof(struct stats_counter);
+
+ BNX2X_PCI_ALLOC(bp->fw_stats, &bp->fw_stats_mapping,
+ bp->fw_stats_data_sz + bp->fw_stats_req_sz);
+
+ /* Set shortcuts */
+ bp->fw_stats_req = (struct bnx2x_fw_stats_req *)bp->fw_stats;
+ bp->fw_stats_req_mapping = bp->fw_stats_mapping;
+ bp->fw_stats_data = (struct bnx2x_fw_stats_data *)
+ ((u8 *)bp->fw_stats + bp->fw_stats_req_sz);
+ bp->fw_stats_data_mapping = bp->fw_stats_mapping +
+ bp->fw_stats_req_sz;
+
+ DP(BNX2X_MSG_SP, "statistics request base address set to %x %x",
+ U64_HI(bp->fw_stats_req_mapping),
+ U64_LO(bp->fw_stats_req_mapping));
+ DP(BNX2X_MSG_SP, "statistics data base address set to %x %x",
+ U64_HI(bp->fw_stats_data_mapping),
+ U64_LO(bp->fw_stats_data_mapping));
+ return 0;
+
+alloc_mem_err:
+ bnx2x_free_fw_stats_mem(bp);
+ BNX2X_ERR("Can't allocate FW stats memory\n");
+ return -ENOMEM;
+}
+
+/* send load request to mcp and analyze response */
+static int bnx2x_nic_load_request(struct bnx2x *bp, u32 *load_code)
+{
+ /* init fw_seq */
+ bp->fw_seq =
+ (SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_mb_header) &
+ DRV_MSG_SEQ_NUMBER_MASK);
+ BNX2X_DEV_INFO("fw_seq 0x%08x\n", bp->fw_seq);
+
+ /* Get current FW pulse sequence */
+ bp->fw_drv_pulse_wr_seq =
+ (SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_pulse_mb) &
+ DRV_PULSE_SEQ_MASK);
+ BNX2X_DEV_INFO("drv_pulse 0x%x\n", bp->fw_drv_pulse_wr_seq);
+
+ /* load request */
+ (*load_code) = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_REQ,
+ DRV_MSG_CODE_LOAD_REQ_WITH_LFA);
+
+ /* if mcp fails to respond we must abort */
+ if (!(*load_code)) {
+ BNX2X_ERR("MCP response failure, aborting\n");
+ return -EBUSY;
+ }
+
+ /* If mcp refused (e.g. other port is in diagnostic mode) we
+ * must abort
+ */
+ if ((*load_code) == FW_MSG_CODE_DRV_LOAD_REFUSED) {
+ BNX2X_ERR("MCP refused load request, aborting\n");
+ return -EBUSY;
+ }
+ return 0;
+}
+
+/* check whether another PF has already loaded FW to chip. In
+ * virtualized environments a pf from another VM may have already
+ * initialized the device including loading FW
+ */
+int bnx2x_nic_load_analyze_req(struct bnx2x *bp, u32 load_code)
+{
+ /* is another pf loaded on this engine? */
+ if (load_code != FW_MSG_CODE_DRV_LOAD_COMMON_CHIP &&
+ load_code != FW_MSG_CODE_DRV_LOAD_COMMON) {
+ /* build my FW version dword */
+ u32 my_fw = (BCM_5710_FW_MAJOR_VERSION) +
+ (BCM_5710_FW_MINOR_VERSION << 8) +
+ (BCM_5710_FW_REVISION_VERSION << 16) +
+ (BCM_5710_FW_ENGINEERING_VERSION << 24);
+
+ /* read loaded FW from chip */
+ u32 loaded_fw = REG_RD(bp, XSEM_REG_PRAM);
+
+ DP(BNX2X_MSG_SP, "loaded fw %x, my fw %x\n",
+ loaded_fw, my_fw);
+
+ /* abort nic load if version mismatch */
+ if (my_fw != loaded_fw) {
+ BNX2X_ERR("bnx2x with FW %x was already loaded which mismatches my %x FW. aborting\n",
loaded_fw, my_fw);
- return false;
+ return -EBUSY;
+ }
+ }
+ return 0;
+}
+
+/* returns the "mcp load_code" according to global load_count array */
+static int bnx2x_nic_load_no_mcp(struct bnx2x *bp, int port)
+{
+ int path = BP_PATH(bp);
+
+ DP(NETIF_MSG_IFUP, "NO MCP - load counts[%d] %d, %d, %d\n",
+ path, load_count[path][0], load_count[path][1],
+ load_count[path][2]);
+ load_count[path][0]++;
+ load_count[path][1 + port]++;
+ DP(NETIF_MSG_IFUP, "NO MCP - new load counts[%d] %d, %d, %d\n",
+ path, load_count[path][0], load_count[path][1],
+ load_count[path][2]);
+ if (load_count[path][0] == 1)
+ return FW_MSG_CODE_DRV_LOAD_COMMON;
+ else if (load_count[path][1 + port] == 1)
+ return FW_MSG_CODE_DRV_LOAD_PORT;
+ else
+ return FW_MSG_CODE_DRV_LOAD_FUNCTION;
+}
+
+/* mark PMF if applicable */
+static void bnx2x_nic_load_pmf(struct bnx2x *bp, u32 load_code)
+{
+ if ((load_code == FW_MSG_CODE_DRV_LOAD_COMMON) ||
+ (load_code == FW_MSG_CODE_DRV_LOAD_COMMON_CHIP) ||
+ (load_code == FW_MSG_CODE_DRV_LOAD_PORT)) {
+ bp->port.pmf = 1;
+ /* We need the barrier to ensure the ordering between the
+ * writing to bp->port.pmf here and reading it from the
+ * bnx2x_periodic_task().
+ */
+ smp_mb();
+ } else {
+ bp->port.pmf = 0;
}
- return true;
+ DP(NETIF_MSG_LINK, "pmf %d\n", bp->port.pmf);
+}
+
+static void bnx2x_nic_load_afex_dcc(struct bnx2x *bp, int load_code)
+{
+ if (((load_code == FW_MSG_CODE_DRV_LOAD_COMMON) ||
+ (load_code == FW_MSG_CODE_DRV_LOAD_COMMON_CHIP)) &&
+ (bp->common.shmem2_base)) {
+ if (SHMEM2_HAS(bp, dcc_support))
+ SHMEM2_WR(bp, dcc_support,
+ (SHMEM_DCC_SUPPORT_DISABLE_ENABLE_PF_TLV |
+ SHMEM_DCC_SUPPORT_BANDWIDTH_ALLOCATION_TLV));
+ if (SHMEM2_HAS(bp, afex_driver_support))
+ SHMEM2_WR(bp, afex_driver_support,
+ SHMEM_AFEX_SUPPORTED_VERSION_ONE);
+ }
+
+ /* Set AFEX default VLAN tag to an invalid value */
+ bp->afex_def_vlan_tag = -1;
}
/**
@@ -2095,10 +2293,12 @@ int bnx2x_load_cnic(struct bnx2x *bp)
mutex_init(&bp->cnic_mutex);
- rc = bnx2x_alloc_mem_cnic(bp);
- if (rc) {
- BNX2X_ERR("Unable to allocate bp memory for cnic\n");
- LOAD_ERROR_EXIT_CNIC(bp, load_error_cnic0);
+ if (IS_PF(bp)) {
+ rc = bnx2x_alloc_mem_cnic(bp);
+ if (rc) {
+ BNX2X_ERR("Unable to allocate bp memory for cnic\n");
+ LOAD_ERROR_EXIT_CNIC(bp, load_error_cnic0);
+ }
}
rc = bnx2x_alloc_fp_mem_cnic(bp);
@@ -2125,14 +2325,17 @@ int bnx2x_load_cnic(struct bnx2x *bp)
bnx2x_nic_init_cnic(bp);
- /* Enable Timer scan */
- REG_WR(bp, TM_REG_EN_LINEAR0_TIMER + port*4, 1);
-
- for_each_cnic_queue(bp, i) {
- rc = bnx2x_setup_queue(bp, &bp->fp[i], 0);
- if (rc) {
- BNX2X_ERR("Queue setup failed\n");
- LOAD_ERROR_EXIT(bp, load_error_cnic2);
+ if (IS_PF(bp)) {
+ /* Enable Timer scan */
+ REG_WR(bp, TM_REG_EN_LINEAR0_TIMER + port*4, 1);
+
+ /* setup cnic queues */
+ for_each_cnic_queue(bp, i) {
+ rc = bnx2x_setup_queue(bp, &bp->fp[i], 0);
+ if (rc) {
+ BNX2X_ERR("Queue setup failed\n");
+ LOAD_ERROR_EXIT(bp, load_error_cnic2);
+ }
}
}
@@ -2178,8 +2381,7 @@ load_error_cnic0:
int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
{
int port = BP_PORT(bp);
- u32 load_code;
- int i, rc;
+ int i, rc = 0, load_code = 0;
DP(NETIF_MSG_IFUP, "Starting NIC load\n");
DP(NETIF_MSG_IFUP,
@@ -2201,8 +2403,9 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
&bp->last_reported_link.link_report_flags);
bnx2x_release_phy_lock(bp);
- /* must be called before memory allocation and HW init */
- bnx2x_ilt_set_info(bp);
+ if (IS_PF(bp))
+ /* must be called before memory allocation and HW init */
+ bnx2x_ilt_set_info(bp);
/*
* Zero fastpath structures preserving invariants like napi, which are
@@ -2221,8 +2424,26 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
/* Set the receive queues buffer size */
bnx2x_set_rx_buf_size(bp);
- if (bnx2x_alloc_mem(bp))
- return -ENOMEM;
+ if (IS_PF(bp)) {
+ rc = bnx2x_alloc_mem(bp);
+ if (rc) {
+ BNX2X_ERR("Unable to allocate bp memory\n");
+ return rc;
+ }
+ }
+
+ /* Allocated memory for FW statistics */
+ if (bnx2x_alloc_fw_stats_mem(bp))
+ LOAD_ERROR_EXIT(bp, load_error0);
+
+ /* need to be done after alloc mem, since it's self adjusting to amount
+ * of memory available for RSS queues
+ */
+ rc = bnx2x_alloc_fp_mem(bp);
+ if (rc) {
+ BNX2X_ERR("Unable to allocate memory for fps\n");
+ LOAD_ERROR_EXIT(bp, load_error0);
+ }
/* As long as bnx2x_alloc_mem() may possibly update
* bp->num_queues, bnx2x_set_real_num_queues() should always
@@ -2245,98 +2466,48 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
DP(NETIF_MSG_IFUP, "napi added\n");
bnx2x_napi_enable(bp);
- /* set pf load just before approaching the MCP */
- bnx2x_set_pf_load(bp);
-
- /* Send LOAD_REQUEST command to MCP
- * Returns the type of LOAD command:
- * if it is the first port to be initialized
- * common blocks should be initialized, otherwise - not
- */
- if (!BP_NOMCP(bp)) {
- /* init fw_seq */
- bp->fw_seq =
- (SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_mb_header) &
- DRV_MSG_SEQ_NUMBER_MASK);
- BNX2X_DEV_INFO("fw_seq 0x%08x\n", bp->fw_seq);
-
- /* Get current FW pulse sequence */
- bp->fw_drv_pulse_wr_seq =
- (SHMEM_RD(bp, func_mb[BP_FW_MB_IDX(bp)].drv_pulse_mb) &
- DRV_PULSE_SEQ_MASK);
- BNX2X_DEV_INFO("drv_pulse 0x%x\n", bp->fw_drv_pulse_wr_seq);
-
- load_code = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_REQ,
- DRV_MSG_CODE_LOAD_REQ_WITH_LFA);
- if (!load_code) {
- BNX2X_ERR("MCP response failure, aborting\n");
- rc = -EBUSY;
- LOAD_ERROR_EXIT(bp, load_error1);
- }
- if (load_code == FW_MSG_CODE_DRV_LOAD_REFUSED) {
- BNX2X_ERR("Driver load refused\n");
- rc = -EBUSY; /* other port in diagnostic mode */
- LOAD_ERROR_EXIT(bp, load_error1);
- }
- if (load_code != FW_MSG_CODE_DRV_LOAD_COMMON_CHIP &&
- load_code != FW_MSG_CODE_DRV_LOAD_COMMON) {
- /* abort nic load if version mismatch */
- if (!bnx2x_test_firmware_version(bp, true)) {
- rc = -EBUSY;
+ if (IS_PF(bp)) {
+ /* set pf load just before approaching the MCP */
+ bnx2x_set_pf_load(bp);
+
+ /* if mcp exists send load request and analyze response */
+ if (!BP_NOMCP(bp)) {
+ /* attempt to load pf */
+ rc = bnx2x_nic_load_request(bp, &load_code);
+ if (rc)
+ LOAD_ERROR_EXIT(bp, load_error1);
+
+ /* what did mcp say? */
+ rc = bnx2x_nic_load_analyze_req(bp, load_code);
+ if (rc) {
+ bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
LOAD_ERROR_EXIT(bp, load_error2);
}
+ } else {
+ load_code = bnx2x_nic_load_no_mcp(bp, port);
}
- } else {
- int path = BP_PATH(bp);
-
- DP(NETIF_MSG_IFUP, "NO MCP - load counts[%d] %d, %d, %d\n",
- path, load_count[path][0], load_count[path][1],
- load_count[path][2]);
- load_count[path][0]++;
- load_count[path][1 + port]++;
- DP(NETIF_MSG_IFUP, "NO MCP - new load counts[%d] %d, %d, %d\n",
- path, load_count[path][0], load_count[path][1],
- load_count[path][2]);
- if (load_count[path][0] == 1)
- load_code = FW_MSG_CODE_DRV_LOAD_COMMON;
- else if (load_count[path][1 + port] == 1)
- load_code = FW_MSG_CODE_DRV_LOAD_PORT;
- else
- load_code = FW_MSG_CODE_DRV_LOAD_FUNCTION;
- }
+ /* mark pmf if applicable */
+ bnx2x_nic_load_pmf(bp, load_code);
- if ((load_code == FW_MSG_CODE_DRV_LOAD_COMMON) ||
- (load_code == FW_MSG_CODE_DRV_LOAD_COMMON_CHIP) ||
- (load_code == FW_MSG_CODE_DRV_LOAD_PORT)) {
- bp->port.pmf = 1;
- /*
- * We need the barrier to ensure the ordering between the
- * writing to bp->port.pmf here and reading it from the
- * bnx2x_periodic_task().
- */
- smp_mb();
- } else
- bp->port.pmf = 0;
-
- DP(NETIF_MSG_IFUP, "pmf %d\n", bp->port.pmf);
+ /* Init Function state controlling object */
+ bnx2x__init_func_obj(bp);
- /* Init Function state controlling object */
- bnx2x__init_func_obj(bp);
-
- /* Initialize HW */
- rc = bnx2x_init_hw(bp, load_code);
- if (rc) {
- BNX2X_ERR("HW init failed, aborting\n");
- bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
- LOAD_ERROR_EXIT(bp, load_error2);
+ /* Initialize HW */
+ rc = bnx2x_init_hw(bp, load_code);
+ if (rc) {
+ BNX2X_ERR("HW init failed, aborting\n");
+ bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
+ LOAD_ERROR_EXIT(bp, load_error2);
+ }
}
/* Connect to IRQs */
rc = bnx2x_setup_irqs(bp);
if (rc) {
- BNX2X_ERR("IRQs setup failed\n");
- bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
+ BNX2X_ERR("setup irqs failed\n");
+ if (IS_PF(bp))
+ bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
LOAD_ERROR_EXIT(bp, load_error2);
}
@@ -2344,78 +2515,78 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
bnx2x_nic_init(bp, load_code);
/* Init per-function objects */
- bnx2x_init_bp_objs(bp);
-
- if (((load_code == FW_MSG_CODE_DRV_LOAD_COMMON) ||
- (load_code == FW_MSG_CODE_DRV_LOAD_COMMON_CHIP)) &&
- (bp->common.shmem2_base)) {
- if (SHMEM2_HAS(bp, dcc_support))
- SHMEM2_WR(bp, dcc_support,
- (SHMEM_DCC_SUPPORT_DISABLE_ENABLE_PF_TLV |
- SHMEM_DCC_SUPPORT_BANDWIDTH_ALLOCATION_TLV));
- if (SHMEM2_HAS(bp, afex_driver_support))
- SHMEM2_WR(bp, afex_driver_support,
- SHMEM_AFEX_SUPPORTED_VERSION_ONE);
- }
+ if (IS_PF(bp)) {
+ bnx2x_init_bp_objs(bp);
- /* Set AFEX default VLAN tag to an invalid value */
- bp->afex_def_vlan_tag = -1;
- bp->state = BNX2X_STATE_OPENING_WAIT4_PORT;
- rc = bnx2x_func_start(bp);
- if (rc) {
- BNX2X_ERR("Function start failed!\n");
- bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
- LOAD_ERROR_EXIT(bp, load_error3);
- }
+ /* Set AFEX default VLAN tag to an invalid value */
+ bp->afex_def_vlan_tag = -1;
+ bnx2x_nic_load_afex_dcc(bp, load_code);
+ bp->state = BNX2X_STATE_OPENING_WAIT4_PORT;
+ rc = bnx2x_func_start(bp);
+ if (rc) {
+ BNX2X_ERR("Function start failed!\n");
+ bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
- /* Send LOAD_DONE command to MCP */
- if (!BP_NOMCP(bp)) {
- load_code = bnx2x_fw_command(bp, DRV_MSG_CODE_LOAD_DONE, 0);
- if (!load_code) {
- BNX2X_ERR("MCP response failure, aborting\n");
- rc = -EBUSY;
LOAD_ERROR_EXIT(bp, load_error3);
}
- }
- rc = bnx2x_setup_leading(bp);
- if (rc) {
- BNX2X_ERR("Setup leading failed!\n");
- LOAD_ERROR_EXIT(bp, load_error3);
- }
+ /* Send LOAD_DONE command to MCP */
+ if (!BP_NOMCP(bp)) {
+ load_code = bnx2x_fw_command(bp,
+ DRV_MSG_CODE_LOAD_DONE, 0);
+ if (!load_code) {
+ BNX2X_ERR("MCP response failure, aborting\n");
+ rc = -EBUSY;
+ LOAD_ERROR_EXIT(bp, load_error3);
+ }
+ }
- for_each_nondefault_eth_queue(bp, i) {
- rc = bnx2x_setup_queue(bp, &bp->fp[i], 0);
+ /* setup the leading queue */
+ rc = bnx2x_setup_leading(bp);
if (rc) {
- BNX2X_ERR("Queue setup failed\n");
+ BNX2X_ERR("Setup leading failed!\n");
LOAD_ERROR_EXIT(bp, load_error3);
}
- }
- rc = bnx2x_init_rss_pf(bp);
- if (rc) {
- BNX2X_ERR("PF RSS init failed\n");
- LOAD_ERROR_EXIT(bp, load_error3);
+ /* set up the rest of the queues */
+ for_each_nondefault_eth_queue(bp, i) {
+ rc = bnx2x_setup_queue(bp, &bp->fp[i], 0);
+ if (rc) {
+ BNX2X_ERR("Queue setup failed\n");
+ LOAD_ERROR_EXIT(bp, load_error3);
+ }
+ }
+
+ /* setup rss */
+ rc = bnx2x_init_rss_pf(bp);
+ if (rc) {
+ BNX2X_ERR("PF RSS init failed\n");
+ LOAD_ERROR_EXIT(bp, load_error3);
+ }
}
/* Now when Clients are configured we are ready to work */
bp->state = BNX2X_STATE_OPEN;
/* Configure a ucast MAC */
- rc = bnx2x_set_eth_mac(bp, true);
+ if (IS_PF(bp))
+ rc = bnx2x_set_eth_mac(bp, true);
if (rc) {
BNX2X_ERR("Setting Ethernet MAC failed\n");
LOAD_ERROR_EXIT(bp, load_error3);
}
- if (bp->pending_max) {
+ if (IS_PF(bp) && bp->pending_max) {
bnx2x_update_max_mf_config(bp, bp->pending_max);
bp->pending_max = 0;
}
- if (bp->port.pmf)
- bnx2x_initial_phy_init(bp, load_mode);
+ if (bp->port.pmf) {
+ rc = bnx2x_initial_phy_init(bp, load_mode);
+ if (rc)
+ LOAD_ERROR_EXIT(bp, load_error3);
+ }
bp->link_params.feature_config_flags &= ~FEATURE_CONFIG_BOOT_FROM_SAN;
/* Start fast path */
@@ -2457,8 +2628,8 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
if (CNIC_ENABLED(bp))
bnx2x_load_cnic(bp);
- /* mark driver is loaded in shmem2 */
- if (SHMEM2_HAS(bp, drv_capabilities_flag)) {
+ if (IS_PF(bp) && SHMEM2_HAS(bp, drv_capabilities_flag)) {
+ /* mark driver is loaded in shmem2 */
u32 val;
val = SHMEM2_RD(bp, drv_capabilities_flag[BP_FW_MB_IDX(bp)]);
SHMEM2_WR(bp, drv_capabilities_flag[BP_FW_MB_IDX(bp)],
@@ -2467,7 +2638,7 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
}
/* Wait for all pending SP commands to complete */
- if (!bnx2x_wait_sp_comp(bp, ~0x0UL)) {
+ if (IS_PF(bp) && !bnx2x_wait_sp_comp(bp, ~0x0UL)) {
BNX2X_ERR("Timeout waiting for SP elements to complete\n");
bnx2x_nic_unload(bp, UNLOAD_CLOSE, false);
return -EBUSY;
@@ -2483,10 +2654,12 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
#ifndef BNX2X_STOP_ON_ERROR
load_error3:
- bnx2x_int_disable_sync(bp, 1);
+ if (IS_PF(bp)) {
+ bnx2x_int_disable_sync(bp, 1);
- /* Clean queueable objects */
- bnx2x_squeeze_objects(bp);
+ /* Clean queueable objects */
+ bnx2x_squeeze_objects(bp);
+ }
/* Free SKBs, SGEs, TPA pool and driver internals */
bnx2x_free_skbs(bp);
@@ -2496,7 +2669,7 @@ load_error3:
/* Release IRQs */
bnx2x_free_irq(bp);
load_error2:
- if (!BP_NOMCP(bp)) {
+ if (IS_PF(bp) && !BP_NOMCP(bp)) {
bnx2x_fw_command(bp, DRV_MSG_CODE_UNLOAD_REQ_WOL_MCP, 0);
bnx2x_fw_command(bp, DRV_MSG_CODE_UNLOAD_DONE, 0);
}
@@ -2504,15 +2677,35 @@ load_error2:
bp->port.pmf = 0;
load_error1:
bnx2x_napi_disable(bp);
+
/* clear pf_load status, as it was already set */
- bnx2x_clear_pf_load(bp);
+ if (IS_PF(bp))
+ bnx2x_clear_pf_load(bp);
load_error0:
+ bnx2x_free_fp_mem(bp);
+ bnx2x_free_fw_stats_mem(bp);
bnx2x_free_mem(bp);
return rc;
#endif /* ! BNX2X_STOP_ON_ERROR */
}
+static int bnx2x_drain_tx_queues(struct bnx2x *bp)
+{
+ u8 rc = 0, cos, i;
+
+ /* Wait until tx fastpath tasks complete */
+ for_each_tx_queue(bp, i) {
+ struct bnx2x_fastpath *fp = &bp->fp[i];
+
+ for_each_cos_in_tx_queue(fp, cos)
+ rc = bnx2x_clean_tx_queue(bp, fp->txdata_ptr[cos]);
+ if (rc)
+ return rc;
+ }
+ return 0;
+}
+
/* must be called with rtnl_lock */
int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
{
@@ -2522,15 +2715,16 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
DP(NETIF_MSG_IFUP, "Starting NIC unload\n");
/* mark driver is unloaded in shmem2 */
- if (SHMEM2_HAS(bp, drv_capabilities_flag)) {
+ if (IS_PF(bp) && SHMEM2_HAS(bp, drv_capabilities_flag)) {
u32 val;
val = SHMEM2_RD(bp, drv_capabilities_flag[BP_FW_MB_IDX(bp)]);
SHMEM2_WR(bp, drv_capabilities_flag[BP_FW_MB_IDX(bp)],
val & ~DRV_FLAGS_CAPABILITIES_LOADED_L2);
}
- if ((bp->state == BNX2X_STATE_CLOSED) ||
- (bp->state == BNX2X_STATE_ERROR)) {
+ if (IS_PF(bp) &&
+ (bp->state == BNX2X_STATE_CLOSED ||
+ bp->state == BNX2X_STATE_ERROR)) {
/* We can get here if the driver has been unloaded
* during parity error recovery and is either waiting for a
* leader to complete or for other functions to unload and
@@ -2567,13 +2761,16 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
del_timer_sync(&bp->timer);
- /* Set ALWAYS_ALIVE bit in shmem */
- bp->fw_drv_pulse_wr_seq |= DRV_PULSE_ALWAYS_ALIVE;
-
- bnx2x_drv_pulse(bp);
+ if (IS_PF(bp)) {
+ /* Set ALWAYS_ALIVE bit in shmem */
+ bp->fw_drv_pulse_wr_seq |= DRV_PULSE_ALWAYS_ALIVE;
+ bnx2x_drv_pulse(bp);
+ bnx2x_stats_handle(bp, STATS_EVENT_STOP);
+ bnx2x_save_statistics(bp);
+ }
- bnx2x_stats_handle(bp, STATS_EVENT_STOP);
- bnx2x_save_statistics(bp);
+ /* wait till consumers catch up with producers in all queues */
+ bnx2x_drain_tx_queues(bp);
/* Cleanup the chip if needed */
if (unload_mode != UNLOAD_RECOVERY)
@@ -2609,7 +2806,8 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
* At this stage no more interrupts will arrive so we may safly clean
* the queueable objects here in case they failed to get cleaned so far.
*/
- bnx2x_squeeze_objects(bp);
+ if (IS_PF(bp))
+ bnx2x_squeeze_objects(bp);
/* There should be no more pending SP commands at this stage */
bp->sp_state = 0;
@@ -2623,19 +2821,22 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
for_each_rx_queue(bp, i)
bnx2x_free_rx_sge_range(bp, bp->fp + i, NUM_RX_SGE);
- if (CNIC_LOADED(bp)) {
+ bnx2x_free_fp_mem(bp);
+ if (CNIC_LOADED(bp))
bnx2x_free_fp_mem_cnic(bp);
- bnx2x_free_mem_cnic(bp);
- }
- bnx2x_free_mem(bp);
+ if (IS_PF(bp)) {
+ bnx2x_free_mem(bp);
+ if (CNIC_LOADED(bp))
+ bnx2x_free_mem_cnic(bp);
+ }
bp->state = BNX2X_STATE_CLOSED;
bp->cnic_loaded = false;
/* Check if there are pending parity attentions. If there are - set
* RECOVERY_IN_PROGRESS.
*/
- if (bnx2x_chk_parity_attn(bp, &global, false)) {
+ if (IS_PF(bp) && bnx2x_chk_parity_attn(bp, &global, false)) {
bnx2x_set_reset_in_progress(bp);
/* Set RESET_IS_GLOBAL if needed */
@@ -2647,7 +2848,9 @@ int bnx2x_nic_unload(struct bnx2x *bp, int unload_mode, bool keep_link)
/* The last driver must disable a "close the gate" if there is no
* parity attention or "process kill" pending.
*/
- if (!bnx2x_clear_pf_load(bp) && bnx2x_reset_is_done(bp, BP_PATH(bp)))
+ if (IS_PF(bp) &&
+ !bnx2x_clear_pf_load(bp) &&
+ bnx2x_reset_is_done(bp, BP_PATH(bp)))
bnx2x_disable_close_the_gate(bp);
DP(NETIF_MSG_IFUP, "Ending NIC unload\n");
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
index bca371e..91e432d 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h
@@ -1128,11 +1128,18 @@ static inline u8 bnx2x_fp_qzone_id(struct bnx2x_fastpath *fp)
static inline u32 bnx2x_rx_ustorm_prods_offset(struct bnx2x_fastpath *fp)
{
struct bnx2x *bp = fp->bp;
-
- if (!CHIP_IS_E1x(bp))
- return USTORM_RX_PRODS_E2_OFFSET(fp->cl_qzone_id);
+ u32 offset = BAR_USTRORM_INTMEM;
+
+ if (IS_VF(bp))
+ return PXP_VF_ADDR_USDM_QUEUES_START +
+ bp->acquire_resp.resc.hw_qid[fp->index] *
+ sizeof(struct ustorm_queue_zone_data);
+ else if (!CHIP_IS_E1x(bp))
+ offset += USTORM_RX_PRODS_E2_OFFSET(fp->cl_qzone_id);
else
- return USTORM_RX_PRODS_E1X_OFFSET(BP_PORT(bp), fp->cl_id);
+ offset += USTORM_RX_PRODS_E1X_OFFSET(BP_PORT(bp), fp->cl_id);
+
+ return offset;
}
static inline void bnx2x_init_txdata(struct bnx2x *bp,
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
index 277f17e..b7c82f9 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
@@ -890,7 +890,7 @@ static void bnx2x_set_msglevel(struct net_device *dev, u32 level)
if (capable(CAP_NET_ADMIN)) {
/* dump MCP trace */
- if (level & BNX2X_MSG_MCP)
+ if (IS_PF(bp) && (level & BNX2X_MSG_MCP))
bnx2x_fw_dump_lvl(bp, KERN_INFO);
bp->msg_enable = level;
}
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 3dd5b26..97b8aa7 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -2460,17 +2460,49 @@ void bnx2x__link_status_update(struct bnx2x *bp)
return;
/* read updated dcb configuration */
- bnx2x_dcbx_pmf_update(bp);
-
- bnx2x_link_status_update(&bp->link_params, &bp->link_vars);
+ if (IS_PF(bp)) {
+ bnx2x_dcbx_pmf_update(bp);
+ bnx2x_link_status_update(&bp->link_params, &bp->link_vars);
+ if (bp->link_vars.link_up)
+ bnx2x_stats_handle(bp, STATS_EVENT_LINK_UP);
+ else
+ bnx2x_stats_handle(bp, STATS_EVENT_STOP);
+ /* indicate link status */
+ bnx2x_link_report(bp);
- if (bp->link_vars.link_up)
+ } else { /* VF */
+ bp->port.supported[0] |= (SUPPORTED_10baseT_Half |
+ SUPPORTED_10baseT_Full |
+ SUPPORTED_100baseT_Half |
+ SUPPORTED_100baseT_Full |
+ SUPPORTED_1000baseT_Full |
+ SUPPORTED_2500baseX_Full |
+ SUPPORTED_10000baseT_Full |
+ SUPPORTED_TP |
+ SUPPORTED_FIBRE |
+ SUPPORTED_Autoneg |
+ SUPPORTED_Pause |
+ SUPPORTED_Asym_Pause);
+ bp->port.advertising[0] = bp->port.supported[0];
+
+ bp->link_params.bp = bp;
+ bp->link_params.port = BP_PORT(bp);
+ bp->link_params.req_duplex[0] = DUPLEX_FULL;
+ bp->link_params.req_flow_ctrl[0] = BNX2X_FLOW_CTRL_NONE;
+ bp->link_params.req_line_speed[0] = SPEED_10000;
+ bp->link_params.speed_cap_mask[0] = 0x7f0000;
+ bp->link_params.switch_cfg = SWITCH_CFG_10G;
+ bp->link_vars.mac_type = MAC_TYPE_BMAC;
+ bp->link_vars.line_speed = SPEED_10000;
+ bp->link_vars.link_status =
+ (LINK_STATUS_LINK_UP |
+ LINK_STATUS_SPEED_AND_DUPLEX_10GTFD);
+ bp->link_vars.link_up = 1;
+ bp->link_vars.duplex = DUPLEX_FULL;
+ bp->link_vars.flow_ctrl = BNX2X_FLOW_CTRL_NONE;
+ __bnx2x_link_report(bp);
bnx2x_stats_handle(bp, STATS_EVENT_LINK_UP);
- else
- bnx2x_stats_handle(bp, STATS_EVENT_STOP);
-
- /* indicate link status */
- bnx2x_link_report(bp);
+ }
}
static int bnx2x_afex_func_update(struct bnx2x *bp, u16 vifid,
@@ -5700,6 +5732,13 @@ static void bnx2x_init_eth_fp(struct bnx2x *bp, int fp_idx)
cids[cos] = fp->txdata_ptr[cos]->cid;
}
+ /* nothing more for vf to do here */
+ if (IS_VF(bp))
+ return;
+
+ bnx2x_init_sb(bp, fp->status_blk_mapping, BNX2X_VF_ID_INVALID, false,
+ fp->fw_sb_id, fp->igu_sb_id);
+ bnx2x_update_fpsb_idx(fp);
bnx2x_init_queue_obj(bp, &bnx2x_sp_obj(bp, fp).q_obj, fp->cl_id, cids,
fp->max_cos, BP_FUNC(bp), bnx2x_sp(bp, q_rdata),
bnx2x_sp_mapping(bp, q_rdata), q_type);
@@ -5709,13 +5748,10 @@ static void bnx2x_init_eth_fp(struct bnx2x *bp, int fp_idx)
*/
bnx2x_init_vlan_mac_fp_objs(fp, BNX2X_OBJ_TYPE_RX_TX);
- DP(NETIF_MSG_IFUP, "queue[%d]: bnx2x_init_sb(%p,%p) cl_id %d fw_sb %d igu_sb %d\n",
- fp_idx, bp, fp->status_blk.e2_sb, fp->cl_id, fp->fw_sb_id,
- fp->igu_sb_id);
- bnx2x_init_sb(bp, fp->status_blk_mapping, BNX2X_VF_ID_INVALID, false,
- fp->fw_sb_id, fp->igu_sb_id);
-
- bnx2x_update_fpsb_idx(fp);
+ DP(NETIF_MSG_IFUP,
+ "queue[%d]: bnx2x_init_sb(%p,%p) cl_id %d fw_sb %d igu_sb %d\n",
+ fp_idx, bp, fp->status_blk.e2_sb, fp->cl_id, fp->fw_sb_id,
+ fp->igu_sb_id);
}
static void bnx2x_init_tx_ring_one(struct bnx2x_fp_txdata *txdata)
@@ -5787,17 +5823,22 @@ void bnx2x_nic_init(struct bnx2x *bp, u32 load_code)
for_each_eth_queue(bp, i)
bnx2x_init_eth_fp(bp, i);
+
+ /* ensure status block indices were read */
+ rmb();
+ bnx2x_init_rx_rings(bp);
+ bnx2x_init_tx_rings(bp);
+
+ if (IS_VF(bp))
+ return;
+
/* Initialize MOD_ABS interrupts */
bnx2x_init_mod_abs_int(bp, &bp->link_vars, bp->common.chip_id,
bp->common.shmem_base, bp->common.shmem2_base,
BP_PORT(bp));
- /* ensure status block indices were read */
- rmb();
bnx2x_init_def_sb(bp);
bnx2x_update_dsb_idx(bp);
- bnx2x_init_rx_rings(bp);
- bnx2x_init_tx_rings(bp);
bnx2x_init_sp_ring(bp);
bnx2x_init_eq_ring(bp);
bnx2x_init_internal(bp, load_code);
@@ -9656,7 +9697,7 @@ static int bnx2x_prev_unload_uncommon(struct bnx2x *bp)
* the one required, then FLR will be sufficient to clean any residue
* left by previous driver
*/
- rc = bnx2x_test_firmware_version(bp, false);
+ rc = bnx2x_nic_load_analyze_req(bp, FW_MSG_CODE_DRV_LOAD_FUNCTION);
if (!rc) {
/* fw version is good */
@@ -11236,17 +11277,21 @@ static int bnx2x_open(struct net_device *dev)
bnx2x_set_power_state(bp, PCI_D0);
- other_load_status = bnx2x_get_load_status(bp, other_engine);
- load_status = bnx2x_get_load_status(bp, BP_PATH(bp));
+ if (IS_PF(bp)) {
+ other_load_status = bnx2x_get_load_status(bp, other_engine);
+ load_status = bnx2x_get_load_status(bp, BP_PATH(bp));
+ }
/*
* If parity had happen during the unload, then attentions
* and/or RECOVERY_IN_PROGRES may still be set. In this case we
* want the first function loaded on the current engine to
* complete the recovery.
+ * Parity recovery is only relevant for PF driver.
*/
- if (!bnx2x_reset_is_done(bp, BP_PATH(bp)) ||
- bnx2x_chk_parity_attn(bp, &global, true))
+ if (IS_PF(bp) &&
+ (!bnx2x_reset_is_done(bp, BP_PATH(bp)) ||
+ bnx2x_chk_parity_attn(bp, &global, true)))
do {
/*
* If there are attentions and they are in a global
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index c302de4..3f01526 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -6559,6 +6559,12 @@
#define PXP_VF_ADDR_IGU_END\
((PXP_VF_ADDR_IGU_START) + (PXP_VF_ADDR_IGU_SIZE) - 1)
+#define PXP_VF_ADDR_USDM_QUEUES_START 0x3000
+#define PXP_VF_ADDR_USDM_QUEUES_SIZE\
+ (PXP_VF_ADRR_NUM_QUEUES * PXP_ADDR_QUEUE_SIZE)
+#define PXP_VF_ADDR_USDM_QUEUES_END\
+ ((PXP_VF_ADDR_USDM_QUEUES_START) + (PXP_VF_ADDR_USDM_QUEUES_SIZE) - 1)
+
#define PXP_VF_ADDR_CSDM_GLOBAL_START 0x7600
#define PXP_VF_ADDR_CSDM_GLOBAL_SIZE (PXP_ADDR_REG_SIZE)
#define PXP_VF_ADDR_CSDM_GLOBAL_END\
--
1.7.9.GIT
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox