RDMA/siw: Release xarray entry

[ Upstream commit a3d83276d98886879b5bf7b30b7c29882754e4df ]

The xarray entry is allocated in siw_qp_add(), but release was
missed in case zero-sized SQ was discovered.

Fixes: 661f385961f0 ("RDMA/siw: Fix handling of zero-sized Read and Receive Queues.")
Link: https://lore.kernel.org/r/f070b59d5a1114d5a4e830346755c2b3f141cde5.1620560472.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

diff --git a/drivers/infiniband/sw/siw/siw_verbs.c b/drivers/infiniband/sw/siw/siw_verbs.c
index 11bd320..34e847a 100644 (file)
--- a/drivers/infiniband/sw/siw/siw_verbs.c
+++ b/drivers/infiniband/sw/siw/siw_verbs.c
@@ -372,7 +372,7 @@ struct ib_qp *siw_create_qp(struct ib_pd *pd,
        else {
                /* Zero sized SQ is not supported */
                rv = -EINVAL;
-               goto err_out;
+               goto err_out_xa;
        }
        if (num_rqe)
                num_rqe = roundup_pow_of_two(num_rqe);