On Mon, Jan 28, 2008 at 11:14:41AM -0800, Ralph Campbell wrote: > Can you send out a specification for what ib_modify_cq() > is supposed to do? It isn't part of the IBTA verbs spec. > so I don't know how other HCAs are supposed to implement it. >
Many NICs provide for controlling the rate of generation of interrupts. Controlling the rate of interrupt generation helps reducing the overhead of interrupt handling. In IB, a CQ may generate notifications which are asynchronous from the software perspective and may generate interrupts. The following interface allows to control this function on the devices which support this. A notification will be generated if any of two conditions become true. 1. cq_count - number of CQEs were pushed into the CQ since the last notificaton. 2. cq_period - number of micorseconds elapsed since the last notification. After the generation of a notification the coditions are reset. > > + * ib_modify_cq - Modifies moderation params of the CQ > > + * @cq: The CQ to modify. > > + * @cq_count: number of CQEs that will trigger an event > > + * @cq_period: max period of time in usec before triggering an event > > + * > > + */ > > +int ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period); > > + > _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
