aboutsummaryrefslogtreecommitdiff
path: root/drivers/infiniband/ulp/ipoib
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/infiniband/ulp/ipoib')
-rw-r--r--drivers/infiniband/ulp/ipoib/Makefile3
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib.h33
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_cm.c78
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_ethtool.c21
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_ib.c105
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_main.c216
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_multicast.c50
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_netlink.c182
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_verbs.c3
-rw-r--r--drivers/infiniband/ulp/ipoib/ipoib_vlan.c132
10 files changed, 604 insertions, 219 deletions
diff --git a/drivers/infiniband/ulp/ipoib/Makefile b/drivers/infiniband/ulp/ipoib/Makefile
index 3090100f0de..e5430dd5076 100644
--- a/drivers/infiniband/ulp/ipoib/Makefile
+++ b/drivers/infiniband/ulp/ipoib/Makefile
@@ -5,7 +5,8 @@ ib_ipoib-y := ipoib_main.o \
ipoib_multicast.o \
ipoib_verbs.o \
ipoib_vlan.o \
- ipoib_ethtool.o
+ ipoib_ethtool.o \
+ ipoib_netlink.o
ib_ipoib-$(CONFIG_INFINIBAND_IPOIB_CM) += ipoib_cm.o
ib_ipoib-$(CONFIG_INFINIBAND_IPOIB_DEBUG) += ipoib_fs.o
diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index ca43901ed86..c639f90cfda 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -101,9 +101,14 @@ enum {
IPOIB_MCAST_FLAG_SENDONLY = 1,
IPOIB_MCAST_FLAG_BUSY = 2, /* joining or already joined */
IPOIB_MCAST_FLAG_ATTACHED = 3,
+ IPOIB_MCAST_JOIN_STARTED = 4,
MAX_SEND_CQE = 16,
IPOIB_CM_COPYBREAK = 256,
+
+ IPOIB_NON_CHILD = 0,
+ IPOIB_LEGACY_CHILD = 1,
+ IPOIB_RTNL_CHILD = 2,
};
#define IPOIB_OP_RECV (1ul << 31)
@@ -113,6 +118,8 @@ enum {
#define IPOIB_OP_CM (0)
#endif
+#define IPOIB_QPN_MASK ((__force u32) cpu_to_be32(0xFFFFFF))
+
/* structs */
struct ipoib_header {
@@ -145,6 +152,7 @@ struct ipoib_mcast {
struct sk_buff_head pkt_queue;
struct net_device *dev;
+ struct completion done;
};
struct ipoib_rx_buf {
@@ -262,7 +270,10 @@ struct ipoib_ethtool_st {
u16 max_coalesced_frames;
};
+struct ipoib_neigh_table;
+
struct ipoib_neigh_hash {
+ struct ipoib_neigh_table *ntbl;
struct ipoib_neigh __rcu **buckets;
struct rcu_head rcu;
u32 mask;
@@ -271,9 +282,9 @@ struct ipoib_neigh_hash {
struct ipoib_neigh_table {
struct ipoib_neigh_hash __rcu *htbl;
- rwlock_t rwlock;
atomic_t entries;
struct completion flushed;
+ struct completion deleted;
};
/*
@@ -290,7 +301,7 @@ struct ipoib_dev_priv {
unsigned long flags;
- struct mutex vlan_mutex;
+ struct rw_semaphore vlan_rwsem;
struct rb_root path_tree;
struct list_head path_list;
@@ -350,6 +361,7 @@ struct ipoib_dev_priv {
struct net_device *parent;
struct list_head child_intfs;
struct list_head list;
+ int child_type;
#ifdef CONFIG_INFINIBAND_IPOIB_CM
struct ipoib_cm_dev_priv cm;
@@ -509,6 +521,17 @@ void ipoib_event(struct ib_event_handler *handler,
int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey);
int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey);
+int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv,
+ u16 pkey, int child_type);
+
+int __init ipoib_netlink_init(void);
+void __exit ipoib_netlink_fini(void);
+
+void ipoib_set_umcast(struct net_device *ndev, int umcast_val);
+int ipoib_set_mode(struct net_device *dev, const char *buf);
+
+void ipoib_setup(struct net_device *dev);
+
void ipoib_pkey_poll(struct work_struct *work);
int ipoib_pkey_dev_delay_open(struct net_device *dev);
void ipoib_drain_cq(struct net_device *dev);
@@ -516,14 +539,14 @@ void ipoib_drain_cq(struct net_device *dev);
void ipoib_set_ethtool_ops(struct net_device *dev);
int ipoib_set_dev_features(struct ipoib_dev_priv *priv, struct ib_device *hca);
-#ifdef CONFIG_INFINIBAND_IPOIB_CM
-
#define IPOIB_FLAGS_RC 0x80
#define IPOIB_FLAGS_UC 0x40
/* We don't support UC connections at the moment */
#define IPOIB_CM_SUPPORTED(ha) (ha[0] & (IPOIB_FLAGS_RC))
+#ifdef CONFIG_INFINIBAND_IPOIB_CM
+
extern int ipoib_max_conn_qp;
static inline int ipoib_cm_admin_enabled(struct net_device *dev)
@@ -741,4 +764,6 @@ extern int ipoib_debug_level;
#define IPOIB_QPN(ha) (be32_to_cpup((__be32 *) ha) & 0xffffff)
+extern const char ipoib_driver_version[];
+
#endif /* _IPOIB_H */
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
index 95ecf4eadf5..933efcea0d0 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
@@ -140,7 +140,8 @@ static int ipoib_cm_post_receive_nonsrq(struct net_device *dev,
static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev,
struct ipoib_cm_rx_buf *rx_ring,
int id, int frags,
- u64 mapping[IPOIB_CM_RX_SG])
+ u64 mapping[IPOIB_CM_RX_SG],
+ gfp_t gfp)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
struct sk_buff *skb;
@@ -164,7 +165,7 @@ static struct sk_buff *ipoib_cm_alloc_rx_skb(struct net_device *dev,
}
for (i = 0; i < frags; i++) {
- struct page *page = alloc_page(GFP_ATOMIC);
+ struct page *page = alloc_page(gfp);
if (!page)
goto partial_error;
@@ -382,7 +383,8 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i
for (i = 0; i < ipoib_recvq_size; ++i) {
if (!ipoib_cm_alloc_rx_skb(dev, rx->rx_ring, i, IPOIB_CM_RX_SG - 1,
- rx->rx_ring[i].mapping)) {
+ rx->rx_ring[i].mapping,
+ GFP_KERNEL)) {
ipoib_warn(priv, "failed to allocate receive buffer %d\n", i);
ret = -ENOMEM;
goto err_count;
@@ -460,7 +462,7 @@ static int ipoib_cm_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *even
goto err_qp;
}
- psn = random32() & 0xffffff;
+ psn = prandom_u32() & 0xffffff;
ret = ipoib_cm_modify_rx_qp(dev, cm_id, p->qp, psn);
if (ret)
goto err_modify;
@@ -639,7 +641,8 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc)
frags = PAGE_ALIGN(wc->byte_len - min(wc->byte_len,
(unsigned)IPOIB_CM_HEAD_SIZE)) / PAGE_SIZE;
- newskb = ipoib_cm_alloc_rx_skb(dev, rx_ring, wr_id, frags, mapping);
+ newskb = ipoib_cm_alloc_rx_skb(dev, rx_ring, wr_id, frags,
+ mapping, GFP_ATOMIC);
if (unlikely(!newskb)) {
/*
* If we can't allocate a new RX buffer, dump
@@ -741,6 +744,9 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_
tx_req->mapping = addr;
+ skb_orphan(skb);
+ skb_dst_drop(skb);
+
rc = post_send(priv, tx, tx->tx_head & (ipoib_sendq_size - 1),
addr, skb->len);
if (unlikely(rc)) {
@@ -755,9 +761,13 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_
if (++priv->tx_outstanding == ipoib_sendq_size) {
ipoib_dbg(priv, "TX ring 0x%x full, stopping kernel net queue\n",
tx->qp->qp_num);
- if (ib_req_notify_cq(priv->send_cq, IB_CQ_NEXT_COMP))
- ipoib_warn(priv, "request notify on send CQ failed\n");
netif_stop_queue(dev);
+ rc = ib_req_notify_cq(priv->send_cq,
+ IB_CQ_NEXT_COMP | IB_CQ_REPORT_MISSED_EVENTS);
+ if (rc < 0)
+ ipoib_warn(priv, "request notify on send CQ failed\n");
+ else if (rc)
+ ipoib_send_comp_handler(priv->send_cq, dev);
}
}
}
@@ -810,7 +820,6 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc)
if (neigh) {
neigh->cm = NULL;
- list_del(&neigh->list);
ipoib_neigh_free(neigh);
tx->neigh = NULL;
@@ -1021,10 +1030,20 @@ static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_
.cap.max_send_sge = 1,
.sq_sig_type = IB_SIGNAL_ALL_WR,
.qp_type = IB_QPT_RC,
- .qp_context = tx
+ .qp_context = tx,
+ .create_flags = IB_QP_CREATE_USE_GFP_NOIO
};
- return ib_create_qp(priv->pd, &attr);
+ struct ib_qp *tx_qp;
+
+ tx_qp = ib_create_qp(priv->pd, &attr);
+ if (PTR_ERR(tx_qp) == -EINVAL) {
+ ipoib_warn(priv, "can't use GFP_NOIO for QPs on device %s, using GFP_KERNEL\n",
+ priv->ca->name);
+ attr.create_flags &= ~IB_QP_CREATE_USE_GFP_NOIO;
+ tx_qp = ib_create_qp(priv->pd, &attr);
+ }
+ return tx_qp;
}
static int ipoib_cm_send_req(struct net_device *dev,
@@ -1095,12 +1114,14 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn,
struct ipoib_dev_priv *priv = netdev_priv(p->dev);
int ret;
- p->tx_ring = vzalloc(ipoib_sendq_size * sizeof *p->tx_ring);
+ p->tx_ring = __vmalloc(ipoib_sendq_size * sizeof *p->tx_ring,
+ GFP_NOIO, PAGE_KERNEL);
if (!p->tx_ring) {
ipoib_warn(priv, "failed to allocate tx ring\n");
ret = -ENOMEM;
goto err_tx;
}
+ memset(p->tx_ring, 0, ipoib_sendq_size * sizeof *p->tx_ring);
p->qp = ipoib_cm_create_tx_qp(p->dev, p);
if (IS_ERR(p->qp)) {
@@ -1227,7 +1248,6 @@ static int ipoib_cm_tx_handler(struct ib_cm_id *cm_id,
if (neigh) {
neigh->cm = NULL;
- list_del(&neigh->list);
ipoib_neigh_free(neigh);
tx->neigh = NULL;
@@ -1271,12 +1291,15 @@ struct ipoib_cm_tx *ipoib_cm_create_tx(struct net_device *dev, struct ipoib_path
void ipoib_cm_destroy_tx(struct ipoib_cm_tx *tx)
{
struct ipoib_dev_priv *priv = netdev_priv(tx->dev);
+ unsigned long flags;
if (test_and_clear_bit(IPOIB_FLAG_INITIALIZED, &tx->flags)) {
+ spin_lock_irqsave(&priv->lock, flags);
list_move(&tx->list, &priv->cm.reap_list);
queue_work(ipoib_workqueue, &priv->cm.reap_task);
ipoib_dbg(priv, "Reap connection for gid %pI6\n",
tx->neigh->daddr + 4);
tx->neigh = NULL;
+ spin_unlock_irqrestore(&priv->lock, flags);
}
}
@@ -1315,7 +1338,6 @@ static void ipoib_cm_tx_start(struct work_struct *work)
neigh = p->neigh;
if (neigh) {
neigh->cm = NULL;
- list_del(&neigh->list);
ipoib_neigh_free(neigh);
}
list_del(&p->list);
@@ -1449,36 +1471,19 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr,
const char *buf, size_t count)
{
struct net_device *dev = to_net_dev(d);
- struct ipoib_dev_priv *priv = netdev_priv(dev);
+ int ret;
if (!rtnl_trylock())
return restart_syscall();
- /* flush paths if we switch modes so that connections are restarted */
- if (IPOIB_CM_SUPPORTED(dev->dev_addr) && !strcmp(buf, "connected\n")) {
- set_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags);
- ipoib_warn(priv, "enabling connected mode "
- "will cause multicast packet drops\n");
- netdev_update_features(dev);
- rtnl_unlock();
- priv->tx_wr.send_flags &= ~IB_SEND_IP_CSUM;
-
- ipoib_flush_paths(dev);
- return count;
- }
+ ret = ipoib_set_mode(dev, buf);
- if (!strcmp(buf, "datagram\n")) {
- clear_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags);
- netdev_update_features(dev);
- dev_set_mtu(dev, min(priv->mcast_mtu, dev->mtu));
- rtnl_unlock();
- ipoib_flush_paths(dev);
+ rtnl_unlock();
+ if (!ret)
return count;
- }
- rtnl_unlock();
- return -EINVAL;
+ return ret;
}
static DEVICE_ATTR(mode, S_IWUSR | S_IRUGO, show_mode, set_mode);
@@ -1566,7 +1571,8 @@ int ipoib_cm_dev_init(struct net_device *dev)
for (i = 0; i < ipoib_recvq_size; ++i) {
if (!ipoib_cm_alloc_rx_skb(dev, priv->cm.srq_ring, i,
priv->cm.num_frags - 1,
- priv->cm.srq_ring[i].mapping)) {
+ priv->cm.srq_ring[i].mapping,
+ GFP_KERNEL)) {
ipoib_warn(priv, "failed to allocate "
"receive buffer %d\n", i);
ipoib_cm_dev_cleanup(dev);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
index 29bc7b5724a..078cadd6c79 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c
@@ -39,7 +39,24 @@
static void ipoib_get_drvinfo(struct net_device *netdev,
struct ethtool_drvinfo *drvinfo)
{
- strncpy(drvinfo->driver, "ipoib", sizeof(drvinfo->driver) - 1);
+ struct ipoib_dev_priv *priv = netdev_priv(netdev);
+ struct ib_device_attr *attr;
+
+ attr = kmalloc(sizeof(*attr), GFP_KERNEL);
+ if (attr && !ib_query_device(priv->ca, attr))
+ snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
+ "%d.%d.%d", (int)(attr->fw_ver >> 32),
+ (int)(attr->fw_ver >> 16) & 0xffff,
+ (int)attr->fw_ver & 0xffff);
+ kfree(attr);
+
+ strlcpy(drvinfo->bus_info, dev_name(priv->ca->dma_device),
+ sizeof(drvinfo->bus_info));
+
+ strlcpy(drvinfo->version, ipoib_driver_version,
+ sizeof(drvinfo->version));
+
+ strlcpy(drvinfo->driver, "ib_ipoib", sizeof(drvinfo->driver));
}
static int ipoib_get_coalesce(struct net_device *dev,
@@ -88,5 +105,5 @@ static const struct ethtool_ops ipoib_ethtool_ops = {
void ipoib_set_ethtool_ops(struct net_device *dev)
{
- SET_ETHTOOL_OPS(dev, &ipoib_ethtool_ops);
+ dev->ethtool_ops = &ipoib_ethtool_ops;
}
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index f10221f4080..6a7003ddb0b 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -600,6 +600,9 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb,
netif_stop_queue(dev);
}
+ skb_orphan(skb);
+ skb_dst_drop(skb);
+
rc = post_send(priv, priv->tx_head & (ipoib_sendq_size - 1),
address->ah, qpn, tx_req, phead, hlen);
if (unlikely(rc)) {
@@ -615,8 +618,6 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb,
address->last_send = priv->tx_head;
++priv->tx_head;
- skb_orphan(skb);
-
}
if (unlikely(priv->tx_outstanding > MAX_SEND_CQE))
@@ -684,15 +685,13 @@ int ipoib_ib_dev_open(struct net_device *dev)
ret = ipoib_ib_post_receives(dev);
if (ret) {
ipoib_warn(priv, "ipoib_ib_post_receives returned %d\n", ret);
- ipoib_ib_dev_stop(dev, 1);
- return -1;
+ goto dev_stop;
}
ret = ipoib_cm_dev_open(dev);
if (ret) {
ipoib_warn(priv, "ipoib_cm_dev_open returned %d\n", ret);
- ipoib_ib_dev_stop(dev, 1);
- return -1;
+ goto dev_stop;
}
clear_bit(IPOIB_STOP_REAPER, &priv->flags);
@@ -703,6 +702,11 @@ int ipoib_ib_dev_open(struct net_device *dev)
napi_enable(&priv->napi);
return 0;
+dev_stop:
+ if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
+ napi_enable(&priv->napi);
+ ipoib_ib_dev_stop(dev, 1);
+ return -1;
}
static void ipoib_pkey_dev_check_presence(struct net_device *dev)
@@ -745,10 +749,8 @@ int ipoib_ib_dev_down(struct net_device *dev, int flush)
if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
mutex_lock(&pkey_mutex);
set_bit(IPOIB_PKEY_STOP, &priv->flags);
- cancel_delayed_work(&priv->pkey_poll_task);
+ cancel_delayed_work_sync(&priv->pkey_poll_task);
mutex_unlock(&pkey_mutex);
- if (flush)
- flush_workqueue(ipoib_workqueue);
}
ipoib_mcast_stop_thread(dev, flush);
@@ -931,14 +933,49 @@ int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port)
return 0;
}
+/*
+ * Takes whatever value which is in pkey index 0 and updates priv->pkey
+ * returns 0 if the pkey value was changed.
+ */
+static inline int update_parent_pkey(struct ipoib_dev_priv *priv)
+{
+ int result;
+ u16 prev_pkey;
+
+ prev_pkey = priv->pkey;
+ result = ib_query_pkey(priv->ca, priv->port, 0, &priv->pkey);
+ if (result) {
+ ipoib_warn(priv, "ib_query_pkey port %d failed (ret = %d)\n",
+ priv->port, result);
+ return result;
+ }
+
+ priv->pkey |= 0x8000;
+
+ if (prev_pkey != priv->pkey) {
+ ipoib_dbg(priv, "pkey changed from 0x%x to 0x%x\n",
+ prev_pkey, priv->pkey);
+ /*
+ * Update the pkey in the broadcast address, while making sure to set
+ * the full membership bit, so that we join the right broadcast group.
+ */
+ priv->dev->broadcast[8] = priv->pkey >> 8;
+ priv->dev->broadcast[9] = priv->pkey & 0xff;
+ return 0;
+ }
+
+ return 1;
+}
+
static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
enum ipoib_flush_level level)
{
struct ipoib_dev_priv *cpriv;
struct net_device *dev = priv->dev;
u16 new_index;
+ int result;
- mutex_lock(&priv->vlan_mutex);
+ down_read(&priv->vlan_rwsem);
/*
* Flush any child interfaces too -- they might be up even if
@@ -947,9 +984,13 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
list_for_each_entry(cpriv, &priv->child_intfs, list)
__ipoib_ib_dev_flush(cpriv, level);
- mutex_unlock(&priv->vlan_mutex);
+ up_read(&priv->vlan_rwsem);
if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) {
+ /* for non-child devices must check/update the pkey value here */
+ if (level == IPOIB_FLUSH_HEAVY &&
+ !test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags))
+ update_parent_pkey(priv);
ipoib_dbg(priv, "Not flushing - IPOIB_FLAG_INITIALIZED not set.\n");
return;
}
@@ -960,21 +1001,32 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
}
if (level == IPOIB_FLUSH_HEAVY) {
- if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) {
- clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
- ipoib_ib_dev_down(dev, 0);
- ipoib_ib_dev_stop(dev, 0);
- if (ipoib_pkey_dev_delay_open(dev))
+ /* child devices chase their origin pkey value, while non-child
+ * (parent) devices should always takes what present in pkey index 0
+ */
+ if (test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) {
+ if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) {
+ clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
+ ipoib_ib_dev_down(dev, 0);
+ ipoib_ib_dev_stop(dev, 0);
+ if (ipoib_pkey_dev_delay_open(dev))
+ return;
+ }
+ /* restart QP only if P_Key index is changed */
+ if (test_and_set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags) &&
+ new_index == priv->pkey_index) {
+ ipoib_dbg(priv, "Not flushing - P_Key index not changed.\n");
return;
+ }
+ priv->pkey_index = new_index;
+ } else {
+ result = update_parent_pkey(priv);
+ /* restart QP only if P_Key value changed */
+ if (result) {
+ ipoib_dbg(priv, "Not flushing - P_Key value not changed.\n");
+ return;
+ }
}
-
- /* restart QP only if P_Key index is changed */
- if (test_and_set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags) &&
- new_index == priv->pkey_index) {
- ipoib_dbg(priv, "Not flushing - P_Key index not changed.\n");
- return;
- }
- priv->pkey_index = new_index;
}
if (level == IPOIB_FLUSH_LIGHT) {
@@ -1030,6 +1082,11 @@ void ipoib_ib_dev_cleanup(struct net_device *dev)
struct ipoib_dev_priv *priv = netdev_priv(dev);
ipoib_dbg(priv, "cleaning up ib_dev\n");
+ /*
+ * We must make sure there are no more (path) completions
+ * that may wish to touch priv fields that are no longer valid
+ */
+ ipoib_flush_paths(dev);
ipoib_mcast_stop_thread(dev, 1);
ipoib_mcast_dev_flush(dev);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 97920b77a5d..5786a78ff8b 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -49,9 +49,14 @@
#include <linux/jhash.h>
#include <net/arp.h>
+#define DRV_VERSION "1.0.0"
+
+const char ipoib_driver_version[] = DRV_VERSION;
+
MODULE_AUTHOR("Roland Dreier");
MODULE_DESCRIPTION("IP-over-InfiniBand net driver");
MODULE_LICENSE("Dual BSD/GPL");
+MODULE_VERSION(DRV_VERSION);
int ipoib_sendq_size __read_mostly = IPOIB_TX_RING_SIZE;
int ipoib_recvq_size __read_mostly = IPOIB_RX_RING_SIZE;
@@ -99,6 +104,8 @@ int ipoib_open(struct net_device *dev)
ipoib_dbg(priv, "bringing up interface\n");
+ netif_carrier_off(dev);
+
set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
if (ipoib_pkey_dev_delay_open(dev))
@@ -114,7 +121,7 @@ int ipoib_open(struct net_device *dev)
struct ipoib_dev_priv *cpriv;
/* Bring up any child interfaces too */
- mutex_lock(&priv->vlan_mutex);
+ down_read(&priv->vlan_rwsem);
list_for_each_entry(cpriv, &priv->child_intfs, list) {
int flags;
@@ -124,7 +131,7 @@ int ipoib_open(struct net_device *dev)
dev_change_flags(cpriv->dev, flags | IFF_UP);
}
- mutex_unlock(&priv->vlan_mutex);
+ up_read(&priv->vlan_rwsem);
}
netif_start_queue(dev);
@@ -150,14 +157,14 @@ static int ipoib_stop(struct net_device *dev)
netif_stop_queue(dev);
- ipoib_ib_dev_down(dev, 0);
+ ipoib_ib_dev_down(dev, 1);
ipoib_ib_dev_stop(dev, 0);
if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) {
struct ipoib_dev_priv *cpriv;
/* Bring down any child interfaces too */
- mutex_lock(&priv->vlan_mutex);
+ down_read(&priv->vlan_rwsem);
list_for_each_entry(cpriv, &priv->child_intfs, list) {
int flags;
@@ -167,12 +174,17 @@ static int ipoib_stop(struct net_device *dev)
dev_change_flags(cpriv->dev, flags & ~IFF_UP);
}
- mutex_unlock(&priv->vlan_mutex);
+ up_read(&priv->vlan_rwsem);
}
return 0;
}
+static void ipoib_uninit(struct net_device *dev)
+{
+ ipoib_dev_cleanup(dev);
+}
+
static netdev_features_t ipoib_fix_features(struct net_device *dev, netdev_features_t features)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
@@ -210,6 +222,37 @@ static int ipoib_change_mtu(struct net_device *dev, int new_mtu)
return 0;
}
+int ipoib_set_mode(struct net_device *dev, const char *buf)
+{
+ struct ipoib_dev_priv *priv = netdev_priv(dev);
+
+ /* flush paths if we switch modes so that connections are restarted */
+ if (IPOIB_CM_SUPPORTED(dev->dev_addr) && !strcmp(buf, "connected\n")) {
+ set_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags);
+ ipoib_warn(priv, "enabling connected mode "
+ "will cause multicast packet drops\n");
+ netdev_update_features(dev);
+ rtnl_unlock();
+ priv->tx_wr.send_flags &= ~IB_SEND_IP_CSUM;
+
+ ipoib_flush_paths(dev);
+ rtnl_lock();
+ return 0;
+ }
+
+ if (!strcmp(buf, "datagram\n")) {
+ clear_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags);
+ netdev_update_features(dev);
+ dev_set_mtu(dev, min(priv->mcast_mtu, dev->mtu));
+ rtnl_unlock();
+ ipoib_flush_paths(dev);
+ rtnl_lock();
+ return 0;
+ }
+
+ return -EINVAL;
+}
+
static struct ipoib_path *__path_find(struct net_device *dev, void *gid)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
@@ -452,7 +495,6 @@ static void path_rec_completion(int status,
path,
neigh));
if (!ipoib_cm_get(neigh)) {
- list_del(&neigh->list);
ipoib_neigh_free(neigh);
continue;
}
@@ -469,6 +511,9 @@ static void path_rec_completion(int status,
spin_unlock_irqrestore(&priv->lock, flags);
+ if (IS_ERR_OR_NULL(ah))
+ ipoib_del_neighs_by_gid(dev, path->pathrec.dgid.raw);
+
if (old_ah)
ipoib_put_ah(old_ah);
@@ -546,15 +591,15 @@ static void neigh_add_path(struct sk_buff *skb, u8 *daddr,
struct ipoib_neigh *neigh;
unsigned long flags;
+ spin_lock_irqsave(&priv->lock, flags);
neigh = ipoib_neigh_alloc(daddr, dev);
if (!neigh) {
+ spin_unlock_irqrestore(&priv->lock, flags);
++dev->stats.tx_dropped;
dev_kfree_skb_any(skb);
return;
}
- spin_lock_irqsave(&priv->lock, flags);
-
path = __path_find(dev, daddr + 4);
if (!path) {
path = path_rec_create(dev, daddr + 4);
@@ -574,7 +619,6 @@ static void neigh_add_path(struct sk_buff *skb, u8 *daddr,
if (!ipoib_cm_get(neigh))
ipoib_cm_set(neigh, ipoib_cm_create_tx(dev, path, neigh));
if (!ipoib_cm_get(neigh)) {
- list_del(&neigh->list);
ipoib_neigh_free(neigh);
goto err_drop;
}
@@ -595,7 +639,7 @@ static void neigh_add_path(struct sk_buff *skb, u8 *daddr,
neigh->ah = NULL;
if (!path->query && path_rec_start(dev, path))
- goto err_list;
+ goto err_path;
__skb_queue_tail(&neigh->queue, skb);
}
@@ -604,9 +648,6 @@ static void neigh_add_path(struct sk_buff *skb, u8 *daddr,
ipoib_neigh_put(neigh);
return;
-err_list:
- list_del(&neigh->list);
-
err_path:
ipoib_neigh_free(neigh);
err_drop:
@@ -686,7 +727,8 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev)
if ((header->proto != htons(ETH_P_IP)) &&
(header->proto != htons(ETH_P_IPV6)) &&
(header->proto != htons(ETH_P_ARP)) &&
- (header->proto != htons(ETH_P_RARP))) {
+ (header->proto != htons(ETH_P_RARP)) &&
+ (header->proto != htons(ETH_P_TIPC))) {
/* ethertype not supported by IPoIB */
++dev->stats.tx_dropped;
dev_kfree_skb_any(skb);
@@ -707,6 +749,7 @@ static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev)
switch (header->proto) {
case htons(ETH_P_IP):
case htons(ETH_P_IPV6):
+ case htons(ETH_P_TIPC):
neigh = ipoib_neigh_get(dev, cb->hwaddr);
if (unlikely(!neigh)) {
neigh_add_path(skb, cb->hwaddr, dev);
@@ -784,7 +827,7 @@ static int ipoib_hard_header(struct sk_buff *skb,
*/
memcpy(cb->hwaddr, daddr, INFINIBAND_ALEN);
- return 0;
+ return sizeof *header;
}
static void ipoib_set_mcast_list(struct net_device *dev)
@@ -808,10 +851,10 @@ static u32 ipoib_addr_hash(struct ipoib_neigh_hash *htbl, u8 *daddr)
* different subnets.
*/
/* qpn octets[1:4) & port GUID octets[12:20) */
- u32 *daddr_32 = (u32 *) daddr;
+ u32 *d32 = (u32 *) daddr;
u32 hv;
- hv = jhash_3words(daddr_32[3], daddr_32[4], 0xFFFFFF & daddr_32[0], 0);
+ hv = jhash_3words(d32[3], d32[4], IPOIB_QPN_MASK & d32[0], 0);
return hv & htbl->mask;
}
@@ -863,10 +906,10 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv)
if (test_bit(IPOIB_STOP_NEIGH_GC, &priv->flags))
return;
- write_lock_bh(&ntbl->rwlock);
+ spin_lock_irqsave(&priv->lock, flags);
htbl = rcu_dereference_protected(ntbl->htbl,
- lockdep_is_held(&ntbl->rwlock));
+ lockdep_is_held(&priv->lock));
if (!htbl)
goto out_unlock;
@@ -883,16 +926,14 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv)
struct ipoib_neigh __rcu **np = &htbl->buckets[i];
while ((neigh = rcu_dereference_protected(*np,
- lockdep_is_held(&ntbl->rwlock))) != NULL) {
+ lockdep_is_held(&priv->lock))) != NULL) {
/* was the neigh idle for two GC periods */
if (time_after(neigh_obsolete, neigh->alive)) {
rcu_assign_pointer(*np,
rcu_dereference_protected(neigh->hnext,
- lockdep_is_held(&ntbl->rwlock)));
+ lockdep_is_held(&priv->lock)));
/* remove from path/mc list */
- spin_lock_irqsave(&priv->lock, flags);
list_del(&neigh->list);
- spin_unlock_irqrestore(&priv->lock, flags);
call_rcu(&neigh->rcu, ipoib_neigh_reclaim);
} else {
np = &neigh->hnext;
@@ -902,7 +943,7 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv)
}
out_unlock:
- write_unlock_bh(&ntbl->rwlock);
+ spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipoib_reap_neigh(struct work_struct *work)
@@ -947,10 +988,8 @@ struct ipoib_neigh *ipoib_neigh_alloc(u8 *daddr,
struct ipoib_neigh *neigh;
u32 hash_val;
- write_lock_bh(&ntbl->rwlock);
-
htbl = rcu_dereference_protected(ntbl->htbl,
- lockdep_is_held(&ntbl->rwlock));
+ lockdep_is_held(&priv->lock));
if (!htbl) {
neigh = NULL;
goto out_unlock;
@@ -961,10 +1000,10 @@ struct ipoib_neigh *ipoib_neigh_alloc(u8 *daddr,
*/
hash_val = ipoib_addr_hash(htbl, daddr);
for (neigh = rcu_dereference_protected(htbl->buckets[hash_val],
- lockdep_is_held(&ntbl->rwlock));
+ lockdep_is_held(&priv->lock));
neigh != NULL;
neigh = rcu_dereference_protected(neigh->hnext,
- lockdep_is_held(&ntbl->rwlock))) {
+ lockdep_is_held(&priv->lock))) {
if (memcmp(daddr, neigh->daddr, INFINIBAND_ALEN) == 0) {
/* found, take one ref on behalf of the caller */
if (!atomic_inc_not_zero(&neigh->refcnt)) {
@@ -987,12 +1026,11 @@ struct ipoib_neigh *ipoib_neigh_alloc(u8 *daddr,
/* put in hash */
rcu_assign_pointer(neigh->hnext,
rcu_dereference_protected(htbl->buckets[hash_val],
- lockdep_is_held(&ntbl->rwlock)));
+ lockdep_is_held(&priv->lock)));
rcu_assign_pointer(htbl->buckets[hash_val], neigh);
atomic_inc(&ntbl->entries);
out_unlock:
- write_unlock_bh(&ntbl->rwlock);
return neigh;
}
@@ -1040,35 +1078,31 @@ void ipoib_neigh_free(struct ipoib_neigh *neigh)
struct ipoib_neigh *n;
u32 hash_val;
- write_lock_bh(&ntbl->rwlock);
-
htbl = rcu_dereference_protected(ntbl->htbl,
- lockdep_is_held(&ntbl->rwlock));
+ lockdep_is_held(&priv->lock));
if (!htbl)
- goto out_unlock;
+ return;
hash_val = ipoib_addr_hash(htbl, neigh->daddr);
np = &htbl->buckets[hash_val];
for (n = rcu_dereference_protected(*np,
- lockdep_is_held(&ntbl->rwlock));
+ lockdep_is_held(&priv->lock));
n != NULL;
- n = rcu_dereference_protected(neigh->hnext,
- lockdep_is_held(&ntbl->rwlock))) {
+ n = rcu_dereference_protected(*np,
+ lockdep_is_held(&priv->lock))) {
if (n == neigh) {
/* found */
rcu_assign_pointer(*np,
rcu_dereference_protected(neigh->hnext,
- lockdep_is_held(&ntbl->rwlock)));
+ lockdep_is_held(&priv->lock)));
+ /* remove from parent list */
+ list_del(&neigh->list);
call_rcu(&neigh->rcu, ipoib_neigh_reclaim);
- goto out_unlock;
+ return;
} else {
np = &n->hnext;
}
}
-
-out_unlock:
- write_unlock_bh(&ntbl->rwlock);
-
}
static int ipoib_neigh_hash_init(struct ipoib_dev_priv *priv)
@@ -1080,7 +1114,6 @@ static int ipoib_neigh_hash_init(struct ipoib_dev_priv *priv)
clear_bit(IPOIB_NEIGH_TBL_FLUSH, &priv->flags);
ntbl->htbl = NULL;
- rwlock_init(&ntbl->rwlock);
htbl = kzalloc(sizeof(*htbl), GFP_KERNEL);
if (!htbl)
return -ENOMEM;
@@ -1095,6 +1128,7 @@ static int ipoib_neigh_hash_init(struct ipoib_dev_priv *priv)
htbl->mask = (size - 1);
htbl->buckets = buckets;
ntbl->htbl = htbl;
+ htbl->ntbl = ntbl;
atomic_set(&ntbl->entries, 0);
/* start garbage collection */
@@ -1111,9 +1145,11 @@ static void neigh_hash_free_rcu(struct rcu_head *head)
struct ipoib_neigh_hash,
rcu);
struct ipoib_neigh __rcu **buckets = htbl->buckets;
+ struct ipoib_neigh_table *ntbl = htbl->ntbl;
kfree(buckets);
kfree(htbl);
+ complete(&ntbl->deleted);
}
void ipoib_del_neighs_by_gid(struct net_device *dev, u8 *gid)
@@ -1125,10 +1161,10 @@ void ipoib_del_neighs_by_gid(struct net_device *dev, u8 *gid)
int i;
/* remove all neigh connected to a given path or mcast */
- write_lock_bh(&ntbl->rwlock);
+ spin_lock_irqsave(&priv->lock, flags);
htbl = rcu_dereference_protected(ntbl->htbl,
- lockdep_is_held(&ntbl->rwlock));
+ lockdep_is_held(&priv->lock));
if (!htbl)
goto out_unlock;
@@ -1138,16 +1174,14 @@ void ipoib_del_neighs_by_gid(struct net_device *dev, u8 *gid)
struct ipoib_neigh __rcu **np = &htbl->buckets[i];
while ((neigh = rcu_dereference_protected(*np,
- lockdep_is_held(&ntbl->rwlock))) != NULL) {
+ lockdep_is_held(&priv->lock))) != NULL) {
/* delete neighs belong to this parent */
if (!memcmp(gid, neigh->daddr + 4, sizeof (union ib_gid))) {
rcu_assign_pointer(*np,
rcu_dereference_protected(neigh->hnext,
- lockdep_is_held(&ntbl->rwlock)));
+ lockdep_is_held(&priv->lock)));
/* remove from parent list */
- spin_lock_irqsave(&priv->lock, flags);
list_del(&neigh->list);
- spin_unlock_irqrestore(&priv->lock, flags);
call_rcu(&neigh->rcu, ipoib_neigh_reclaim);
} else {
np = &neigh->hnext;
@@ -1156,7 +1190,7 @@ void ipoib_del_neighs_by_gid(struct net_device *dev, u8 *gid)
}
}
out_unlock:
- write_unlock_bh(&ntbl->rwlock);
+ spin_unlock_irqrestore(&priv->lock, flags);
}
static void ipoib_flush_neighs(struct ipoib_dev_priv *priv)
@@ -1164,37 +1198,44 @@ static void ipoib_flush_neighs(struct ipoib_dev_priv *priv)
struct ipoib_neigh_table *ntbl = &priv->ntbl;
struct ipoib_neigh_hash *htbl;
unsigned long flags;
- int i;
+ int i, wait_flushed = 0;
- write_lock_bh(&ntbl->rwlock);
+ init_completion(&priv->ntbl.flushed);
+
+ spin_lock_irqsave(&priv->lock, flags);
htbl = rcu_dereference_protected(ntbl->htbl,
- lockdep_is_held(&ntbl->rwlock));
+ lockdep_is_held(&priv->lock));
if (!htbl)
goto out_unlock;
+ wait_flushed = atomic_read(&priv->ntbl.entries);
+ if (!wait_flushed)
+ goto free_htbl;
+
for (i = 0; i < htbl->size; i++) {
struct ipoib_neigh *neigh;
struct ipoib_neigh __rcu **np = &htbl->buckets[i];
while ((neigh = rcu_dereference_protected(*np,
- lockdep_is_held(&ntbl->rwlock))) != NULL) {
+ lockdep_is_held(&priv->lock))) != NULL) {
rcu_assign_pointer(*np,
rcu_dereference_protected(neigh->hnext,
- lockdep_is_held(&ntbl->rwlock)));
+ lockdep_is_held(&priv->lock)));
/* remove from path/mc list */
- spin_lock_irqsave(&priv->lock, flags);
list_del(&neigh->list);
- spin_unlock_irqrestore(&priv->lock, flags);
call_rcu(&neigh->rcu, ipoib_neigh_reclaim);
}
}
+free_htbl:
rcu_assign_pointer(ntbl->htbl, NULL);
call_rcu(&htbl->rcu, neigh_hash_free_rcu);
out_unlock:
- write_unlock_bh(&ntbl->rwlock);
+ spin_unlock_irqrestore(&priv->lock, flags);
+ if (wait_flushed)
+ wait_for_completion(&priv->ntbl.flushed);
}
static void ipoib_neigh_hash_uninit(struct net_device *dev)
@@ -1203,7 +1244,7 @@ static void ipoib_neigh_hash_uninit(struct net_device *dev)
int stopped;
ipoib_dbg(priv, "ipoib_neigh_hash_uninit\n");
- init_completion(&priv->ntbl.flushed);
+ init_completion(&priv->ntbl.deleted);
set_bit(IPOIB_NEIGH_TBL_FLUSH, &priv->flags);
/* Stop GC if called at init fail need to cancel work */
@@ -1211,10 +1252,9 @@ static void ipoib_neigh_hash_uninit(struct net_device *dev)
if (!stopped)
cancel_delayed_work(&priv->neigh_reap_task);
- if (atomic_read(&priv->ntbl.entries)) {
- ipoib_flush_neighs(priv);
- wait_for_completion(&priv->ntbl.flushed);
- }
+ ipoib_flush_neighs(priv);
+
+ wait_for_completion(&priv->ntbl.deleted);
}
@@ -1262,6 +1302,9 @@ out:
void ipoib_dev_cleanup(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev), *cpriv, *tcpriv;
+ LIST_HEAD(head);
+
+ ASSERT_RTNL();
ipoib_delete_debug_files(dev);
@@ -1270,10 +1313,9 @@ void ipoib_dev_cleanup(struct net_device *dev)
/* Stop GC on child */
set_bit(IPOIB_STOP_NEIGH_GC, &cpriv->flags);
cancel_delayed_work(&cpriv->neigh_reap_task);
- unregister_netdev(cpriv->dev);
- ipoib_dev_cleanup(cpriv->dev);
- free_netdev(cpriv->dev);
+ unregister_netdevice_queue(cpriv->dev, &head);
}
+ unregister_netdevice_many(&head);
ipoib_ib_dev_cleanup(dev);
@@ -1291,6 +1333,7 @@ static const struct header_ops ipoib_header_ops = {
};
static const struct net_device_ops ipoib_netdev_ops = {
+ .ndo_uninit = ipoib_uninit,
.ndo_open = ipoib_open,
.ndo_stop = ipoib_stop,
.ndo_change_mtu = ipoib_change_mtu,
@@ -1300,7 +1343,7 @@ static const struct net_device_ops ipoib_netdev_ops = {
.ndo_set_rx_mode = ipoib_set_mcast_list,
};
-static void ipoib_setup(struct net_device *dev)
+void ipoib_setup(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
@@ -1309,7 +1352,7 @@ static void ipoib_setup(struct net_device *dev)
ipoib_set_ethtool_ops(dev);
- netif_napi_add(dev, &priv->napi, ipoib_poll, 100);
+ netif_napi_add(dev, &priv->napi, ipoib_poll, NAPI_POLL_WEIGHT);
dev->watchdog_timeo = HZ;
@@ -1325,13 +1368,11 @@ static void ipoib_setup(struct net_device *dev)
memcpy(dev->broadcast, ipv4_bcast_addr, INFINIBAND_ALEN);
- netif_carrier_off(dev);
-
priv->dev = dev;
spin_lock_init(&priv->lock);
- mutex_init(&priv->vlan_mutex);
+ init_rwsem(&priv->vlan_rwsem);
INIT_LIST_HEAD(&priv->path_list);
INIT_LIST_HEAD(&priv->child_intfs);
@@ -1378,12 +1419,9 @@ static ssize_t show_umcast(struct device *dev,
return sprintf(buf, "%d\n", test_bit(IPOIB_FLAG_UMCAST, &priv->flags));
}
-static ssize_t set_umcast(struct device *dev,
- struct device_attribute *attr,
- const char *buf, size_t count)
+void ipoib_set_umcast(struct net_device *ndev, int umcast_val)
{
- struct ipoib_dev_priv *priv = netdev_priv(to_net_dev(dev));
- unsigned long umcast_val = simple_strtoul(buf, NULL, 0);
+ struct ipoib_dev_priv *priv = netdev_priv(ndev);
if (umcast_val > 0) {
set_bit(IPOIB_FLAG_UMCAST, &priv->flags);
@@ -1391,6 +1429,15 @@ static ssize_t set_umcast(struct device *dev,
"by userspace\n");
} else
clear_bit(IPOIB_FLAG_UMCAST, &priv->flags);
+}
+
+static ssize_t set_umcast(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ unsigned long umcast_val = simple_strtoul(buf, NULL, 0);
+
+ ipoib_set_umcast(to_net_dev(dev), umcast_val);
return count;
}
@@ -1411,7 +1458,7 @@ static ssize_t create_child(struct device *dev,
if (sscanf(buf, "%i", &pkey) != 1)
return -EINVAL;
- if (pkey < 0 || pkey > 0xffff)
+ if (pkey <= 0 || pkey > 0xffff || pkey == 0x8000)
return -EINVAL;
/*
@@ -1648,6 +1695,8 @@ static void ipoib_remove_one(struct ib_device *device)
return;
dev_list = ib_get_client_data(device, &ipoib_client);
+ if (!dev_list)
+ return;
list_for_each_entry_safe(priv, tmp, dev_list, list) {
ib_unregister_event_handler(&priv->event_handler);
@@ -1662,7 +1711,6 @@ static void ipoib_remove_one(struct ib_device *device)
flush_workqueue(ipoib_workqueue);
unregister_netdev(priv->dev);
- ipoib_dev_cleanup(priv->dev);
free_netdev(priv->dev);
}
@@ -1714,8 +1762,15 @@ static int __init ipoib_init_module(void)
if (ret)
goto err_sa;
+ ret = ipoib_netlink_init();
+ if (ret)
+ goto err_client;
+
return 0;
+err_client:
+ ib_unregister_client(&ipoib_client);
+
err_sa:
ib_sa_unregister_client(&ipoib_sa_client);
destroy_workqueue(ipoib_workqueue);
@@ -1728,6 +1783,7 @@ err_fs:
static void __exit ipoib_cleanup_module(void)
{
+ ipoib_netlink_fini();
ib_unregister_client(&ipoib_client);
ib_sa_unregister_client(&ipoib_sa_client);
ipoib_unregister_debugfs();
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 13f4aa7593c..d4e005720d0 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -175,7 +175,9 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast,
mcast->mcmember = *mcmember;
- /* Set the cached Q_Key before we attach if it's the broadcast group */
+ /* Set the multicast MTU and cached Q_Key before we attach if it's
+ * the broadcast group.
+ */
if (!memcmp(mcast->mcmember.mgid.raw, priv->dev->broadcast + 4,
sizeof (union ib_gid))) {
spin_lock_irq(&priv->lock);
@@ -183,10 +185,17 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast,
spin_unlock_irq(&priv->lock);
return -EAGAIN;
}
+ priv->mcast_mtu = IPOIB_UD_MTU(ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu));
priv->qkey = be32_to_cpu(priv->broadcast->mcmember.qkey);
spin_unlock_irq(&priv->lock);
priv->tx_wr.wr.ud.remote_qkey = priv->qkey;
set_qkey = 1;
+
+ if (!ipoib_cm_admin_enabled(dev)) {
+ rtnl_lock();
+ dev_set_mtu(dev, min(priv->mcast_mtu, priv->admin_mtu));
+ rtnl_unlock();
+ }
}
if (!test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) {
@@ -377,8 +386,10 @@ static int ipoib_mcast_join_complete(int status,
mcast->mcmember.mgid.raw, status);
/* We trap for port events ourselves. */
- if (status == -ENETRESET)
- return 0;
+ if (status == -ENETRESET) {
+ status = 0;
+ goto out;
+ }
if (!status)
status = ipoib_mcast_join_finish(mcast, &multicast->rec);
@@ -398,7 +409,8 @@ static int ipoib_mcast_join_complete(int status,
if (mcast == priv->broadcast)
queue_work(ipoib_workqueue, &priv->carrier_on_task);
- return 0;
+ status = 0;
+ goto out;
}
if (mcast->logcount++ < 20) {
@@ -425,7 +437,8 @@ static int ipoib_mcast_join_complete(int status,
mcast->backoff * HZ);
spin_unlock_irq(&priv->lock);
mutex_unlock(&mcast_mutex);
-
+out:
+ complete(&mcast->done);
return status;
}
@@ -475,11 +488,15 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast,
}
set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+ init_completion(&mcast->done);
+ set_bit(IPOIB_MCAST_JOIN_STARTED, &mcast->flags);
+
mcast->mc = ib_sa_join_multicast(&ipoib_sa_client, priv->ca, priv->port,
&rec, comp_mask, GFP_KERNEL,
ipoib_mcast_join_complete, mcast);
if (IS_ERR(mcast->mc)) {
clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+ complete(&mcast->done);
ret = PTR_ERR(mcast->mc);
ipoib_warn(priv, "ib_sa_join_multicast failed, status %d\n", ret);
@@ -501,10 +518,18 @@ void ipoib_mcast_join_task(struct work_struct *work)
struct ipoib_dev_priv *priv =
container_of(work, struct ipoib_dev_priv, mcast_task.work);
struct net_device *dev = priv->dev;
+ struct ib_port_attr port_attr;
if (!test_bit(IPOIB_MCAST_RUN, &priv->flags))
return;
+ if (ib_query_port(priv->ca, priv->port, &port_attr) ||
+ port_attr.state != IB_PORT_ACTIVE) {
+ ipoib_dbg(priv, "port state is not ACTIVE (state = %d) suspending join task\n",
+ port_attr.state);
+ return;
+ }
+
if (ib_query_gid(priv->ca, priv->port, 0, &priv->local_gid))
ipoib_warn(priv, "ib_query_gid() failed\n");
else
@@ -574,14 +599,6 @@ void ipoib_mcast_join_task(struct work_struct *work)
return;
}
- priv->mcast_mtu = IPOIB_UD_MTU(ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu));
-
- if (!ipoib_cm_admin_enabled(dev)) {
- rtnl_lock();
- dev_set_mtu(dev, min(priv->mcast_mtu, priv->admin_mtu));
- rtnl_unlock();
- }
-
ipoib_dbg_mcast(priv, "successfully joined all multicast groups\n");
clear_bit(IPOIB_MCAST_RUN, &priv->flags);
@@ -707,9 +724,7 @@ out:
neigh = ipoib_neigh_get(dev, daddr);
spin_lock_irqsave(&priv->lock, flags);
if (!neigh) {
- spin_unlock_irqrestore(&priv->lock, flags);
neigh = ipoib_neigh_alloc(daddr, dev);
- spin_lock_irqsave(&priv->lock, flags);
if (neigh) {
kref_get(&mcast->ah->ref);
neigh->ah = mcast->ah;
@@ -752,6 +767,11 @@ void ipoib_mcast_dev_flush(struct net_device *dev)
spin_unlock_irqrestore(&priv->lock, flags);
+ /* seperate between the wait to the leave*/
+ list_for_each_entry_safe(mcast, tmcast, &remove_list, list)
+ if (test_bit(IPOIB_MCAST_JOIN_STARTED, &mcast->flags))
+ wait_for_completion(&mcast->done);
+
list_for_each_entry_safe(mcast, tmcast, &remove_list, list) {
ipoib_mcast_leave(dev, mcast);
ipoib_mcast_free(mcast);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_netlink.c b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
new file mode 100644
index 00000000000..cdc7df4fdb8
--- /dev/null
+++ b/drivers/infiniband/ulp/ipoib/ipoib_netlink.c
@@ -0,0 +1,182 @@
+/*
+ * Copyright (c) 2012 Mellanox Technologies. - All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/netdevice.h>
+#include <linux/if_arp.h> /* For ARPHRD_xxx */
+#include <linux/module.h>
+#include <net/rtnetlink.h>
+#include "ipoib.h"
+
+static const struct nla_policy ipoib_policy[IFLA_IPOIB_MAX + 1] = {
+ [IFLA_IPOIB_PKEY] = { .type = NLA_U16 },
+ [IFLA_IPOIB_MODE] = { .type = NLA_U16 },
+ [IFLA_IPOIB_UMCAST] = { .type = NLA_U16 },
+};
+
+static int ipoib_fill_info(struct sk_buff *skb, const struct net_device *dev)
+{
+ struct ipoib_dev_priv *priv = netdev_priv(dev);
+ u16 val;
+
+ if (nla_put_u16(skb, IFLA_IPOIB_PKEY, priv->pkey))
+ goto nla_put_failure;
+
+ val = test_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags);
+ if (nla_put_u16(skb, IFLA_IPOIB_MODE, val))
+ goto nla_put_failure;
+
+ val = test_bit(IPOIB_FLAG_UMCAST, &priv->flags);
+ if (nla_put_u16(skb, IFLA_IPOIB_UMCAST, val))
+ goto nla_put_failure;
+
+ return 0;
+
+nla_put_failure:
+ return -EMSGSIZE;
+}
+
+static int ipoib_changelink(struct net_device *dev,
+ struct nlattr *tb[], struct nlattr *data[])
+{
+ u16 mode, umcast;
+ int ret = 0;
+
+ if (data[IFLA_IPOIB_MODE]) {
+ mode = nla_get_u16(data[IFLA_IPOIB_MODE]);
+ if (mode == IPOIB_MODE_DATAGRAM)
+ ret = ipoib_set_mode(dev, "datagram\n");
+ else if (mode == IPOIB_MODE_CONNECTED)
+ ret = ipoib_set_mode(dev, "connected\n");
+ else
+ ret = -EINVAL;
+
+ if (ret < 0)
+ goto out_err;
+ }
+
+ if (data[IFLA_IPOIB_UMCAST]) {
+ umcast = nla_get_u16(data[IFLA_IPOIB_UMCAST]);
+ ipoib_set_umcast(dev, umcast);
+ }
+
+out_err:
+ return ret;
+}
+
+static int ipoib_new_child_link(struct net *src_net, struct net_device *dev,
+ struct nlattr *tb[], struct nlattr *data[])
+{
+ struct net_device *pdev;
+ struct ipoib_dev_priv *ppriv;
+ u16 child_pkey;
+ int err;
+
+ if (!tb[IFLA_LINK])
+ return -EINVAL;
+
+ pdev = __dev_get_by_index(src_net, nla_get_u32(tb[IFLA_LINK]));
+ if (!pdev || pdev->type != ARPHRD_INFINIBAND)
+ return -ENODEV;
+
+ ppriv = netdev_priv(pdev);
+
+ if (test_bit(IPOIB_FLAG_SUBINTERFACE, &ppriv->flags)) {
+ ipoib_warn(ppriv, "child creation disallowed for child devices\n");
+ return -EINVAL;
+ }
+
+ if (!data || !data[IFLA_IPOIB_PKEY]) {
+ ipoib_dbg(ppriv, "no pkey specified, using parent pkey\n");
+ child_pkey = ppriv->pkey;
+ } else
+ child_pkey = nla_get_u16(data[IFLA_IPOIB_PKEY]);
+
+ if (child_pkey == 0 || child_pkey == 0x8000)
+ return -EINVAL;
+
+ /*
+ * Set the full membership bit, so that we join the right
+ * broadcast group, etc.
+ */
+ child_pkey |= 0x8000;
+
+ err = __ipoib_vlan_add(ppriv, netdev_priv(dev), child_pkey, IPOIB_RTNL_CHILD);
+
+ if (!err && data)
+ err = ipoib_changelink(dev, tb, data);
+ return err;
+}
+
+static void ipoib_unregister_child_dev(struct net_device *dev, struct list_head *head)
+{
+ struct ipoib_dev_priv *priv, *ppriv;
+
+ priv = netdev_priv(dev);
+ ppriv = netdev_priv(priv->parent);
+
+ down_write(&ppriv->vlan_rwsem);
+ unregister_netdevice_queue(dev, head);
+ list_del(&priv->list);
+ up_write(&ppriv->vlan_rwsem);
+}
+
+static size_t ipoib_get_size(const struct net_device *dev)
+{
+ return nla_total_size(2) + /* IFLA_IPOIB_PKEY */
+ nla_total_size(2) + /* IFLA_IPOIB_MODE */
+ nla_total_size(2); /* IFLA_IPOIB_UMCAST */
+}
+
+static struct rtnl_link_ops ipoib_link_ops __read_mostly = {
+ .kind = "ipoib",
+ .maxtype = IFLA_IPOIB_MAX,
+ .policy = ipoib_policy,
+ .priv_size = sizeof(struct ipoib_dev_priv),
+ .setup = ipoib_setup,
+ .newlink = ipoib_new_child_link,
+ .changelink = ipoib_changelink,
+ .dellink = ipoib_unregister_child_dev,
+ .get_size = ipoib_get_size,
+ .fill_info = ipoib_fill_info,
+};
+
+int __init ipoib_netlink_init(void)
+{
+ return rtnl_link_register(&ipoib_link_ops);
+}
+
+void __exit ipoib_netlink_fini(void)
+{
+ rtnl_link_unregister(&ipoib_link_ops);
+}
+
+MODULE_ALIAS_RTNL_LINK("ipoib");
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
index 049a997caff..c56d5d44c53 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
@@ -192,6 +192,9 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca)
if (priv->hca_caps & IB_DEVICE_BLOCK_MULTICAST_LOOPBACK)
init_attr.create_flags |= IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK;
+ if (priv->hca_caps & IB_DEVICE_MANAGED_FLOW_STEERING)
+ init_attr.create_flags |= IB_QP_CREATE_NETIF_QP;
+
if (dev->features & NETIF_F_SG)
init_attr.cap.max_send_sge = MAX_SKB_FRAGS + 1;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
index d7e9740c724..9fad7b5ac8b 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
@@ -49,47 +49,11 @@ static ssize_t show_parent(struct device *d, struct device_attribute *attr,
}
static DEVICE_ATTR(parent, S_IRUGO, show_parent, NULL);
-int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
+int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv,
+ u16 pkey, int type)
{
- struct ipoib_dev_priv *ppriv, *priv;
- char intf_name[IFNAMSIZ];
int result;
- if (!capable(CAP_NET_ADMIN))
- return -EPERM;
-
- ppriv = netdev_priv(pdev);
-
- if (!rtnl_trylock())
- return restart_syscall();
- mutex_lock(&ppriv->vlan_mutex);
-
- /*
- * First ensure this isn't a duplicate. We check the parent device and
- * then all of the child interfaces to make sure the Pkey doesn't match.
- */
- if (ppriv->pkey == pkey) {
- result = -ENOTUNIQ;
- priv = NULL;
- goto err;
- }
-
- list_for_each_entry(priv, &ppriv->child_intfs, list) {
- if (priv->pkey == pkey) {
- result = -ENOTUNIQ;
- priv = NULL;
- goto err;
- }
- }
-
- snprintf(intf_name, sizeof intf_name, "%s.%04x",
- ppriv->dev->name, pkey);
- priv = ipoib_intf_alloc(intf_name);
- if (!priv) {
- result = -ENOMEM;
- goto err;
- }
-
priv->max_ib_mtu = ppriv->max_ib_mtu;
/* MTU will be reset when mcast join happens */
priv->dev->mtu = IPOIB_UD_MTU(priv->max_ib_mtu);
@@ -124,24 +88,27 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
ipoib_create_debug_files(priv->dev);
- if (ipoib_cm_add_mode_attr(priv->dev))
- goto sysfs_failed;
- if (ipoib_add_pkey_attr(priv->dev))
- goto sysfs_failed;
- if (ipoib_add_umcast_attr(priv->dev))
- goto sysfs_failed;
-
- if (device_create_file(&priv->dev->dev, &dev_attr_parent))
- goto sysfs_failed;
+ /* RTNL childs don't need proprietary sysfs entries */
+ if (type == IPOIB_LEGACY_CHILD) {
+ if (ipoib_cm_add_mode_attr(priv->dev))
+ goto sysfs_failed;
+ if (ipoib_add_pkey_attr(priv->dev))
+ goto sysfs_failed;
+ if (ipoib_add_umcast_attr(priv->dev))
+ goto sysfs_failed;
+
+ if (device_create_file(&priv->dev->dev, &dev_attr_parent))
+ goto sysfs_failed;
+ }
+ priv->child_type = type;
+ priv->dev->iflink = ppriv->dev->ifindex;
list_add_tail(&priv->list, &ppriv->child_intfs);
- mutex_unlock(&ppriv->vlan_mutex);
- rtnl_unlock();
-
return 0;
sysfs_failed:
+ result = -ENOMEM;
ipoib_delete_debug_files(priv->dev);
unregister_netdevice(priv->dev);
@@ -149,11 +116,60 @@ register_failed:
ipoib_dev_cleanup(priv->dev);
err:
- mutex_unlock(&ppriv->vlan_mutex);
- rtnl_unlock();
- if (priv)
+ return result;
+}
+
+int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey)
+{
+ struct ipoib_dev_priv *ppriv, *priv;
+ char intf_name[IFNAMSIZ];
+ struct ipoib_dev_priv *tpriv;
+ int result;
+
+ if (!capable(CAP_NET_ADMIN))
+ return -EPERM;
+
+ ppriv = netdev_priv(pdev);
+
+ snprintf(intf_name, sizeof intf_name, "%s.%04x",
+ ppriv->dev->name, pkey);
+ priv = ipoib_intf_alloc(intf_name);
+ if (!priv)
+ return -ENOMEM;
+
+ if (!rtnl_trylock())
+ return restart_syscall();
+
+ down_write(&ppriv->vlan_rwsem);
+
+ /*
+ * First ensure this isn't a duplicate. We check the parent device and
+ * then all of the legacy child interfaces to make sure the Pkey
+ * doesn't match.
+ */
+ if (ppriv->pkey == pkey) {
+ result = -ENOTUNIQ;
+ goto out;
+ }
+
+ list_for_each_entry(tpriv, &ppriv->child_intfs, list) {
+ if (tpriv->pkey == pkey &&
+ tpriv->child_type == IPOIB_LEGACY_CHILD) {
+ result = -ENOTUNIQ;
+ goto out;
+ }
+ }
+
+ result = __ipoib_vlan_add(ppriv, priv, pkey, IPOIB_LEGACY_CHILD);
+
+out:
+ up_write(&ppriv->vlan_rwsem);
+
+ if (result)
free_netdev(priv->dev);
+ rtnl_unlock();
+
return result;
}
@@ -169,17 +185,19 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey)
if (!rtnl_trylock())
return restart_syscall();
- mutex_lock(&ppriv->vlan_mutex);
+
+ down_write(&ppriv->vlan_rwsem);
list_for_each_entry_safe(priv, tpriv, &ppriv->child_intfs, list) {
- if (priv->pkey == pkey) {
+ if (priv->pkey == pkey &&
+ priv->child_type == IPOIB_LEGACY_CHILD) {
unregister_netdevice(priv->dev);
- ipoib_dev_cleanup(priv->dev);
list_del(&priv->list);
dev = priv->dev;
break;
}
}
- mutex_unlock(&ppriv->vlan_mutex);
+ up_write(&ppriv->vlan_rwsem);
+
rtnl_unlock();
if (dev) {
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-12-21 01:55:30 +0000