From: Jim Foraker <[email protected]> If a MAD is sent directly to the local HCA rather than placed on a QP and the MAD fails mkey checks, there is no means to generate a timeout for the client, which may hang. Instead we report IB_MAD_RESULT_FAILURE, which operates the same for on-the-wire packets, but will generate a send failure back to the client.
Reviewed-by: Mike Marciniszyn <[email protected]> Signed-off-by: Jim Foraker <[email protected]> --- drivers/infiniband/hw/qib/qib_mad.c | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/drivers/infiniband/hw/qib/qib_mad.c b/drivers/infiniband/hw/qib/qib_mad.c index ed611e8..4339021 100644 --- a/drivers/infiniband/hw/qib/qib_mad.c +++ b/drivers/infiniband/hw/qib/qib_mad.c @@ -433,7 +433,7 @@ static int check_mkey(struct qib_ibport *ibp, struct ib_smp *smp, int mad_flags) ibp->mkey_lease_period * HZ; /* Generate a trap notice. */ qib_bad_mkey(ibp, smp); - ret = IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_CONSUMED; + ret = 1; } } @@ -464,6 +464,7 @@ static int subn_get_portinfo(struct ib_smp *smp, struct ib_device *ibdev, ibp = to_iport(ibdev, port_num); ret = check_mkey(ibp, smp, 0); if (ret) + ret = IB_MAD_RESULT_FAILURE; goto bail; } } @@ -1848,6 +1849,7 @@ static int process_subn(struct ib_device *ibdev, int mad_flags, port_num && port_num <= ibdev->phys_port_cnt && port != port_num) (void) check_mkey(to_iport(ibdev, port_num), smp, 0); + ret = IB_MAD_RESULT_FAILURE; goto bail; } -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
