diff mbox series

[v6,4/4] RDMA/cma: Avoid GID lookups on iWARP devices

Message ID 168960675257.3007.4737911174148394395.stgit@manet.1015granger.net (mailing list archive)
State Accepted
Delegated to: Jason Gunthorpe
Headers show
Series Handle ARPHRD_NONE devices for siw | expand

Commit Message

Chuck Lever July 17, 2023, 3:12 p.m. UTC
From: Chuck Lever <chuck.lever@oracle.com>

We would like to enable the use of siw on top of a VPN that is
constructed and managed via a tun device. That hasn't worked up
until now because ARPHRD_NONE devices (such as tun devices) have
no GID for the RDMA/core to look up.

But it turns out that the egress device has already been picked for
us -- no GID is necessary. addr_handler() just has to do the right
thing with it.

Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 drivers/infiniband/core/cma.c |   21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)
diff mbox series

Patch

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index da54167723d6..8bd6cb867381 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -700,6 +700,27 @@  cma_validate_port(struct ib_device *device, u32 port,
 	if ((dev_type != ARPHRD_INFINIBAND) && rdma_protocol_ib(device, port))
 		goto out;
 
+	/*
+	 * For drivers that do not associate more than one net device with
+	 * their gid tables, such as iWARP drivers, it is sufficient to
+	 * return the first table entry.
+	 *
+	 * Other driver classes might be included in the future.
+	 */
+	if (rdma_protocol_iwarp(device, port)) {
+		sgid_attr = rdma_get_gid_attr(device, port, 0);
+		if (IS_ERR(sgid_attr))
+			goto out;
+
+		rcu_read_lock();
+		ndev = rcu_dereference(sgid_attr->ndev);
+		if (!net_eq(dev_net(ndev), dev_addr->net) ||
+		    ndev->ifindex != bound_if_index)
+			sgid_attr = ERR_PTR(-ENODEV);
+		rcu_read_unlock();
+		goto out;
+	}
+
 	if (dev_type == ARPHRD_ETHER && rdma_protocol_roce(device, port)) {
 		ndev = dev_get_by_index(dev_addr->net, bound_if_index);
 		if (!ndev)