4.4-stable review patch. If anyone has any objections, please let me know.
------------------ From: Matan Barak <[email protected]> commit 649367735ee5dedb128d9fac0b86ba7e0fe7ae3b upstream. cma_validate_port wrongly assumed that Ethernet devices are RoCE devices and thus their ndev should be matched in the GID table. This broke the iWarp support. Fixing that matching the ndev only if we work on a RoCE port. Cc: <[email protected]> # 4.4.x- Fixes: abae1b71dd37 ('IB/cma: cma_validate_port should verify the port and netdevice') Reported-by: Hariprasad Shenai <[email protected]> Tested-by: Hariprasad Shenai <[email protected]> Signed-off-by: Matan Barak <[email protected]> Reviewed-by: Steve Wise <[email protected]> Signed-off-by: Doug Ledford <[email protected]> Signed-off-by: Steve Wise <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]> --- drivers/infiniband/core/cma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -453,7 +453,7 @@ static inline int cma_validate_port(stru if ((dev_type != ARPHRD_INFINIBAND) && rdma_protocol_ib(device, port)) return ret; - if (dev_type == ARPHRD_ETHER) + if (dev_type == ARPHRD_ETHER && rdma_protocol_roce(device, port)) ndev = dev_get_by_index(&init_net, bound_if_index); ret = ib_find_cached_gid_by_port(device, gid, port, ndev, NULL);
