>> But IPoIB can't really implement NAPI since it's sending work to
>> a shared HCA.
>> And any form of interrupt coalescing would interfere with
>> any latency sensitive work as well (if present).
>Surely. It would have to be configureable for people (like me) that
wanted it.
Configurable feature is not a bad idea for people who are more interested in IPoIB throughput. As I recalled Roland had an experimental of NAPI, it didn't gain too much performance in the previous stack.
Thanks
Shirley Ma
IBM Linux Technology Center
15300 SW Koll Parkway
Beaverton, OR 97006-6063
Phone(Fax): (503) 578-7638
_______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
