On Fri, Jun 22, 2018 at 08:19:15AM -0700, Bart Van Assche wrote:
> Every function that returns COMPST_ERROR must set wqe->status to
> another value than IB_WC_SUCCESS before returning COMPST_ERROR. Fix
> the only code path for which this is not yet the case.
>
> Signed-off-by: Bart Van Assche <bart.vanassche@xxxxxxx>
> Cc: Zhu Yanjun <yanjun.zhu@xxxxxxxxxx>
> Cc: Jianchao Wang <jianchao.w.wang@xxxxxxxxxx>
> Cc: Yuval Shaia <yuval.shaia@xxxxxxxxxx>
> Cc: <stable@xxxxxxxxxxxxxxx>
> ---
> drivers/infiniband/sw/rxe/rxe_comp.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/infiniband/sw/rxe/rxe_comp.c b/drivers/infiniband/sw/rxe/rxe_comp.c
> index 98d470d1f3fc..5bc62c0cc802 100644
> --- a/drivers/infiniband/sw/rxe/rxe_comp.c
> +++ b/drivers/infiniband/sw/rxe/rxe_comp.c
> @@ -276,6 +276,9 @@ static inline enum comp_state check_ack(struct rxe_qp *qp,
> case IB_OPCODE_RC_RDMA_READ_RESPONSE_MIDDLE:
> if (wqe->wr.opcode != IB_WR_RDMA_READ &&
> wqe->wr.opcode != IB_WR_RDMA_READ_WITH_INV) {
> + pr_err("%s: opcode wr %d <> pkt %d\n", __func__,
> + wqe->wr.opcode, pkt->opcode);
> + wqe->status = IB_WC_FATAL_ERR;
Can the user trigger this print? Seems likely.. Probably best not to
have it.
Jason
