[ofa-general] [PATCH RFC v2] IB/ipoib: enable IGMP for userpsace multicast IB apps
Or Gerlitz
ogerlitz at voltaire.com
Thu Sep 20 06:16:50 PDT 2007
changes from v1 - http://lists.openfabrics.org/pipermail/general/2007-September/040250.html
- added module param to control the umcast bit in the device priv flags
- changed the umcast bit name to IPOIB_FLAG_ADMIN_UMCAST_ALLOWED
- the sysfs attribute has now values 0 and 1 instead of "allowed" and "disallowed"
please review and consider for merge to 2.6.24
-----
The kernel IB stack allows (through the RDMA CM) user space multicast applications
to interoperate with IP based apps optionally running at a different IP subnet.
To support this inter-op for the case where the receiving party resides at
the IB side, there is a need to handle IGMP (reports/queries) else the local
IP router would not forward multicast traffic towards the IB network.
This patch does a lookup on the database used for multicast reference counting and
enhances IPoIB to ignore mulicast group which is already handled by user space, all
this under a per device policy flag. That is when the policy flag allows it, IPoIB
will not join and attach its QP to a multicast group which has an entry on the database.
For each IPoIB device, the /sys/class/net/$dev/umcast attribute controls the
policy flag where the default value follows the umcast_allowed module param
(whose default value is zero). The flag can be read and set/unset through sysfs.
Signed-off-by: Or Gerlitz <ogerlitz at voltaire.com>
Index: linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
Index: linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
===================================================================
--- linux-2.6.23-rc5.orig/drivers/infiniband/ulp/ipoib/ipoib_multicast.c 2007-09-20 11:44:58.000000000 +0300
+++ linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib_multicast.c 2007-09-20 12:03:24.000000000 +0300
@@ -783,6 +783,7 @@ void ipoib_mcast_restart_task(struct wor
struct ipoib_mcast *mcast, *tmcast;
LIST_HEAD(remove_list);
unsigned long flags;
+ struct ib_sa_mcmember_rec rec;
ipoib_dbg_mcast(priv, "restarting multicast task\n");
@@ -816,6 +817,15 @@ void ipoib_mcast_restart_task(struct wor
if (!mcast || test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) {
struct ipoib_mcast *nmcast;
+ /* ignore group which is directly joined by user space */
+ if (test_bit(IPOIB_FLAG_ADMIN_UMCAST_ALLOWED, &priv->flags) &&
+ !ib_sa_get_mcmember_rec(priv->ca, priv->port, &mgid, &rec))
+ {
+ ipoib_dbg_mcast(priv, "ignoring multicast entry for mgid "
+ IPOIB_GID_FMT "\n", IPOIB_GID_ARG(mgid));
+ continue;
+ }
+
/* Not found or send-only group, let's add a new entry */
ipoib_dbg_mcast(priv, "adding multicast entry for mgid "
IPOIB_GID_FMT "\n", IPOIB_GID_ARG(mgid));
Index: linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib.h
===================================================================
--- linux-2.6.23-rc5.orig/drivers/infiniband/ulp/ipoib/ipoib.h 2007-09-20 11:44:58.000000000 +0300
+++ linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib.h 2007-09-20 11:49:44.000000000 +0300
@@ -86,6 +86,7 @@ enum {
IPOIB_MCAST_STARTED = 8,
IPOIB_FLAG_NETIF_STOPPED = 9,
IPOIB_FLAG_ADMIN_CM = 10,
+ IPOIB_FLAG_ADMIN_UMCAST_ALLOWED = 11,
IPOIB_MAX_BACKOFF_SECONDS = 16,
@@ -364,6 +365,7 @@ static inline void ipoib_put_ah(struct i
int ipoib_open(struct net_device *dev);
int ipoib_add_pkey_attr(struct net_device *dev);
+int ipoib_add_umcast_attr(struct net_device *dev);
void ipoib_send(struct net_device *dev, struct sk_buff *skb,
struct ipoib_ah *address, u32 qpn);
Index: linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib_main.c
===================================================================
--- linux-2.6.23-rc5.orig/drivers/infiniband/ulp/ipoib/ipoib_main.c 2007-09-20 11:44:58.000000000 +0300
+++ linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib_main.c 2007-09-20 11:55:54.000000000 +0300
@@ -61,6 +61,10 @@ MODULE_PARM_DESC(send_queue_size, "Numbe
module_param_named(recv_queue_size, ipoib_recvq_size, int, 0444);
MODULE_PARM_DESC(recv_queue_size, "Number of descriptors in receive queue");
+int ipoib_umcast_allowed = 0;
+module_param_named(umcast_allowed, ipoib_umcast_allowed, int, 0444);
+MODULE_PARM_DESC(umcast_allowed, "allow ignoring mulicast group which is already handled by user space");
+
#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG
int ipoib_debug_level;
@@ -901,6 +905,9 @@ int ipoib_dev_init(struct net_device *de
if (ipoib_ib_dev_init(dev, ca, port))
goto out_tx_ring_cleanup;
+ if (ipoib_umcast_allowed)
+ set_bit(IPOIB_FLAG_ADMIN_UMCAST_ALLOWED, &priv->flags);
+
return 0;
out_tx_ring_cleanup:
@@ -1017,6 +1024,44 @@ static ssize_t show_pkey(struct device *
}
static DEVICE_ATTR(pkey, S_IRUGO, show_pkey, NULL);
+static ssize_t show_umcast(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct ipoib_dev_priv *priv = netdev_priv(to_net_dev(dev));
+
+ if (test_bit(IPOIB_FLAG_ADMIN_UMCAST_ALLOWED, &priv->flags))
+ return sprintf(buf, "1\n");
+ else
+ return sprintf(buf, "0\n");
+}
+
+static ssize_t set_umcast(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct ipoib_dev_priv *priv = netdev_priv(to_net_dev(dev));
+
+ if (!strcmp(buf, "1\n")) {
+ set_bit(IPOIB_FLAG_ADMIN_UMCAST_ALLOWED, &priv->flags);
+ ipoib_warn(priv, "ignoring multicast groups joined directly "
+ "by user space\n");
+ return count;
+ }
+
+ if (!strcmp(buf, "0\n")) {
+ clear_bit(IPOIB_FLAG_ADMIN_UMCAST_ALLOWED, &priv->flags);
+ return count;
+ }
+
+ return -EINVAL;
+}
+static DEVICE_ATTR(umcast, S_IWUSR | S_IRUGO, show_umcast, set_umcast);
+
+int ipoib_add_umcast_attr(struct net_device *dev)
+{
+ return device_create_file(&dev->dev, &dev_attr_umcast);
+}
+
static ssize_t create_child(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
@@ -1134,6 +1179,8 @@ static struct net_device *ipoib_add_port
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_create_child))
goto sysfs_failed;
if (device_create_file(&priv->dev->dev, &dev_attr_delete_child))
Index: linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib_vlan.c
===================================================================
--- linux-2.6.23-rc5.orig/drivers/infiniband/ulp/ipoib/ipoib_vlan.c 2007-09-20 11:44:58.000000000 +0300
+++ linux-2.6.23-rc5/drivers/infiniband/ulp/ipoib/ipoib_vlan.c 2007-09-20 11:45:46.000000000 +0300
@@ -119,6 +119,8 @@ int ipoib_vlan_add(struct net_device *pd
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;
More information about the general
mailing list