On Thu, 2006-12-14 at 19:20, Sean Hefty wrote:
One minor comment/question below:
> diff --git a/drivers/infiniband/core/notice.c
> b/drivers/infiniband/core/notice.c
> new file mode 100644
> index 0000000..038878d
> --- /dev/null
> +++ b/drivers/infiniband/core/notice.c
[snip...]
> +static int send_unreg(struct inform_group *group)
> +{
> + struct inform_port *port = group->port;
> + struct ib_sa_inform inform;
> + int ret;
> +
> + memset(&inform, 0, sizeof inform);
> + inform.lid_range_begin = cpu_to_be16(0xFFFF);
> + inform.is_generic = 1;
> + inform.type = cpu_to_be16(IB_SA_EVENT_TYPE_ALL);
> + inform.trap.generic.trap_num = cpu_to_be16(group->trap_number);
> + inform.trap.generic.qpn = IB_QP1;
Is the QPN constrained to be QP1 here ?
-- Hal
[snip...]
_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
