Quoting r. Roland Dreier <[EMAIL PROTECTED]>: > Subject: Re: Dropping NETIF_F_SG since no checksum feature. > > Michael> Maybe I can patch linux to allow SG without checksum? > Michael> Dave, maybe you could drop a hint or two on whether this > Michael> is worthwhile and what are the issues that need > Michael> addressing to make this work? > > What do you really gain by allowing SG without checksum? Someone has > to do the checksum anyway, so I don't see that much difference between > doing it in the networking core before passing the data to/from the > driver, or down in the driver itself.
My guess was, an extra pass over data is likely to be expensive - dirtying the cache if nothing else. But I do plan to measure that, and see. -- MST _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
