[PATCH rdma-rc] RDMA/mlx5: Block FDB rules when not in switchdev mode

From: Leon Romanovsky <leon@kernel.org>
To: Doug Ledford <dledford@redhat.com>, Jason Gunthorpe <jgg@nvidia.com>
Cc: Mark Bloch <mbloch@nvidia.com>,
	linux-kernel@vger.kernel.org, linux-rdma@vger.kernel.org,
	Maor Gottlieb <maorg@mellanox.com>,
	Maor Gottlieb <maorg@nvidia.com>
Subject: [PATCH rdma-rc] RDMA/mlx5: Block FDB rules when not in switchdev mode
Date: Mon,  7 Jun 2021 11:03:12 +0300	[thread overview]
Message-ID: <e928ae7c58d07f104716a2a8d730963d1bd01204.1623052923.git.leonro@nvidia.com> (raw)

From: Mark Bloch <mbloch@nvidia.com>

Allow creating FDB steering rules only when in switchdev mode.

The only software model where a userspace application can manipulate
FDB entries is when it manages the eswitch. This is only possible in
switchdev mode where we expose a single RDMA device with representors
for all the vports that are connected to the eswitch.

Fixes: 52438be44112 ("RDMA/mlx5: Allow inserting a steering rule to the FDB")
Reviewed-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
 drivers/infiniband/hw/mlx5/fs.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/infiniband/hw/mlx5/fs.c b/drivers/infiniband/hw/mlx5/fs.c
index 2fc6a60c4e77..f84441ff0c81 100644
--- a/drivers/infiniband/hw/mlx5/fs.c
+++ b/drivers/infiniband/hw/mlx5/fs.c
@@ -2134,6 +2134,12 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_FLOW_MATCHER_CREATE)(
 	if (err)
 		goto end;
 
+	if (obj->ns_type == MLX5_FLOW_NAMESPACE_FDB &&
+	    mlx5_eswitch_mode(dev->mdev) != MLX5_ESWITCH_OFFLOADS) {
+		err = -EINVAL;
+		goto end;
+	}
+
 	uobj->object = obj;
 	obj->mdev = dev->mdev;
 	atomic_set(&obj->usecnt, 0);
-- 
2.31.1

