On Thu, Jul 05, 2018 at 09:41:36AM -0400, Hal Rosenstock wrote:
> On 7/4/2018 8:57 AM, Leon Romanovsky wrote:
> > From: Jason Gunthorpe <jgg@xxxxxxxxxxxx>
> >
> > grh_required is intended to be a global setting where all AV's will require
> > a GRH, not just the sm_lid. Move the special logic to the creation of the
> > SM AH.
> >
> > Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxxxx>
> > Signed-off-by: Artemy Kovalyov <artemyko@xxxxxxxxxxxx>
> > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxxxx>
> > ---
> > drivers/infiniband/core/sa_query.c | 28 ++++++++++++++++++----------
> > drivers/infiniband/hw/hfi1/verbs.c | 9 ---------
> > 2 files changed, 18 insertions(+), 19 deletions(-)
> >
> > diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
> > index 7005afb8a712..0ba09486b09c 100644
> > --- a/drivers/infiniband/core/sa_query.c
> > +++ b/drivers/infiniband/core/sa_query.c
> > @@ -2328,16 +2328,24 @@ static void update_sm_ah(struct work_struct *work)
> > rdma_ah_set_dlid(&ah_attr, port_attr.sm_lid);
> > rdma_ah_set_sl(&ah_attr, port_attr.sm_sl);
> > rdma_ah_set_port_num(&ah_attr, port->port_num);
> > - if (port_attr.grh_required) {
> > - if (ah_attr.type == RDMA_AH_ATTR_TYPE_OPA) {
> > - rdma_ah_set_make_grd(&ah_attr, true);
> > - } else {
> > - rdma_ah_set_ah_flags(&ah_attr, IB_AH_GRH);
> > - rdma_ah_set_subnet_prefix(&ah_attr,
> > - cpu_to_be64(port_attr.subnet_prefix));
> > - rdma_ah_set_interface_id(&ah_attr,
> > - cpu_to_be64(IB_SA_WELL_KNOWN_GUID));
> > - }
> > +
> > + /*
> > + * sm_lid of 0xFFFF needs special handling so that it can be
>
> Would this be better as "OPA sm_lid ..." ?
>
> > + * differentiated from a permissve LID of 0xFFFF. We set the
>
> typo: permissve -> permissive
Thanks Hal,
I'll wait for more feedback and will resend if Jason/Doug decides don't
fix it locally.
>
> > + * grh_required flag here so the SA can program the DGID in the
> > + * address handle appropriately
> > + */
> > + if (ah_attr.type == RDMA_AH_ATTR_TYPE_OPA &&
> > + (port_attr.grh_required ||
> > + port_attr.sm_lid == be16_to_cpu(IB_LID_PERMISSIVE)))
> > + rdma_ah_set_make_grd(&ah_attr, true);
> > +
> > + if (ah_attr.type == RDMA_AH_ATTR_TYPE_IB && port_attr.grh_required) {
> > + rdma_ah_set_ah_flags(&ah_attr, IB_AH_GRH);
> > + rdma_ah_set_subnet_prefix(&ah_attr,
> > + cpu_to_be64(port_attr.subnet_prefix));
> > + rdma_ah_set_interface_id(&ah_attr,
> > + cpu_to_be64(IB_SA_WELL_KNOWN_GUID));
> > }
> >
> > new_ah->ah = rdma_create_ah(port->agent->qp->pd, &ah_attr);
> > diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c
> > index 5cef1224fa9c..2b07a5667ec8 100644
> > --- a/drivers/infiniband/hw/hfi1/verbs.c
> > +++ b/drivers/infiniband/hw/hfi1/verbs.c
> > @@ -1498,15 +1498,6 @@ static int query_port(struct rvt_dev_info *rdi, u8 port_num,
> > props->active_mtu = !valid_ib_mtu(ppd->ibmtu) ? props->max_mtu :
> > mtu_to_enum(ppd->ibmtu, IB_MTU_4096);
> >
> > - /*
> > - * sm_lid of 0xFFFF needs special handling so that it can
> > - * be differentiated from a permissve LID of 0xFFFF.
> > - * We set the grh_required flag here so the SA can program
> > - * the DGID in the address handle appropriately
> > - */
> > - if (props->sm_lid == be16_to_cpu(IB_LID_PERMISSIVE))
> > - props->grh_required = true;
> > -
> > return 0;
> > }
> >
> >
Attachment:
signature.asc
Description: PGP signature
