When calculating the required size of an RC QP send queue, leave enough space for masked atomic operation (which requires more space than "regular" atomic operation).
v2: the wrong git tree was used to create the v1 patch (altough the previous patch can be applied as well without any errors) Signed-off-by: Dotan Barak <dot...@dev.mellanox.co.il> Reviewed-by: Eli Cohen <e...@mellanox.co.il> --- diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 3a91d9d..d91acbf 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -283,7 +283,7 @@ static int send_wqe_overhead(enum ib_qp_type type, u32 flags) sizeof (struct mlx4_wqe_raddr_seg); case IB_QPT_RC: return sizeof (struct mlx4_wqe_ctrl_seg) + - sizeof (struct mlx4_wqe_atomic_seg) + + sizeof (struct mlx4_wqe_masked_atomic_seg) + sizeof (struct mlx4_wqe_raddr_seg); case IB_QPT_SMI: case IB_QPT_GSI: -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html