RDMA experts,

I'd like input on the patch below. iWARP devices that support both native stack TCP and iwarp connections on the same interface need the fix below or some similar enhancement to the rdma cm. This is a bug in the ofed-1.2 RDMA-CM code as it stands. I propose we fix this for ofed-1.2.1 or ofed-1.3.

Here is the issue:

Consider an mpi cluster running mvapich2. And the cluster runs MPI/Sockets jobs concurrently with MPI/RDMA jobs. It is possible, without the patch below, for MPI/Sockets processes to mistakenly get incoming RDMA connections and vice versa. The way mvapich2 works is that the ranks all bind and listen to a random port (retrying new random ports if the bind fails with "in use"). Once they get a free port and bind/listen, they advertise that port to the peers to do connection setup. Currently, without the patch below, the mpi/rdma processes can end up binding/listening to the _same_ port number as the mpi/sockets processes running over the native tcp stack. This is due to duplicate port spaces for native stack TCP and the rdma cm's RDMA_PS_TCP port space. If this happens, then the connections can get screwed up.

The correct solution in my mind is to use the host stack's TCP port space for _all_ RDMA_PS_TCP port allocations. The patch below is a minimal delta to unify the port spaces bay using the kernel stack to bind ports. This is done by allocating a kernel socket and binding to the appropriate local addr/port. It also allows the kernel stack to pick ephemeral ports by virtue of just passing in port 0 on the kernel bind operation.

I'd like to discuss this with the RDMA folks first and iron out an agreement on how this should be implemented, then widen the audience to lklm/netdev. With a goal of inclusion in 2.6.23 and ofed-1.2.1 or 1.3.



-------- Original Message --------
Subject: [PATCH RFC] RDMA/CMA: Allocate PS_TCP ports from the host TCP port space.
Date: Sun, 29 Jul 2007 15:17:04 -0500
From: Steve Wise <[EMAIL PROTECTED]>

RDMA/CMA: Allocate PS_TCP ports from the host TCP port space.

This is needed for iwarp providers that support native and rdma
connections over the same interface.

Signed-off-by: Steve Wise <[EMAIL PROTECTED]>

 drivers/infiniband/core/cma.c |   27 ++++++++++++++++++++++++++-
 1 files changed, 26 insertions(+), 1 deletions(-)

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 9e0ab04..e4d2d7f 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -111,6 +111,7 @@ struct rdma_id_private {
        struct rdma_cm_id       id;

        struct rdma_bind_list   *bind_list;
+       struct socket           *sock;
        struct hlist_node       node;
        struct list_head        list;
        struct list_head        listen_list;
@@ -695,6 +696,8 @@ static void cma_release_port(struct rdma
+       if (id_priv->sock)
+               sock_release(id_priv->sock);

 void rdma_destroy_id(struct rdma_cm_id *id)
@@ -1790,6 +1793,25 @@ static int cma_use_port(struct idr *ps,
        return 0;

+static int cma_get_tcp_port(struct rdma_id_private *id_priv)
+       int ret;
+       struct socket *sock;
+       ret = sock_create_kern(AF_INET, SOCK_STREAM, IPPROTO_TCP, &sock);
+       if (ret)
+               return ret;
+       ret = sock->ops->bind(sock,
+                         (struct socketaddr *)&id_priv->id.route.addr.src_addr,
+                         ip_addr_size(&id_priv->id.route.addr.src_addr));
+       if (ret) {
+               sock_release(sock);
+               return ret;
+       }
+       id_priv->sock = sock;
+       return 0;       
 static int cma_get_port(struct rdma_id_private *id_priv)
        struct idr *ps;
@@ -1801,6 +1823,9 @@ static int cma_get_port(struct rdma_id_p
        case RDMA_PS_TCP:
                ps = &tcp_ps;
+               ret = cma_get_tcp_port(id_priv); /* Synch with native stack */
+               if (ret)
+                       goto out;
        case RDMA_PS_UDP:
                ps = &udp_ps;
@@ -1815,7 +1840,7 @@ static int cma_get_port(struct rdma_id_p
                ret = cma_use_port(ps, id_priv);
        return ret;

general mailing list

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to