3.16.50-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: "Ismail, Mustafa" <mustafa.ism...@intel.com>

commit 5a7a88f1b488e4ee49eb3d5b82612d4d9ffdf2c3 upstream.

The port number is only valid if IB_QP_PORT is set in the mask.
So only check port number if it is valid to prevent modify_qp from
failing due to an invalid port number.

Fixes: 5ecce4c9b17b("Check port number supplied by user verbs cmds")
Reviewed-by: Steve Wise <sw...@opengridcomputing.com>
Signed-off-by: Mustafa Ismail <mustafa.ism...@intel.com>
Tested-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Doug Ledford <dledf...@redhat.com>
[bwh: Backported to 3.16: command structure is cmd not cmd->base]
Signed-off-by: Ben Hutchings <b...@decadent.org.uk>
---
 drivers/infiniband/core/uverbs_cmd.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -1915,7 +1915,8 @@ ssize_t ib_uverbs_modify_qp(struct ib_uv
                goto out;
        }
 
-       if (!rdma_is_port_valid(qp->device, cmd.port_num)) {
+       if ((cmd.attr_mask & IB_QP_PORT) &&
+           !rdma_is_port_valid(qp->device, cmd.port_num)) {
                ret = -EINVAL;
                goto release_qp;
        }

Reply via email to