On Fri, 2005-04-29 at 18:09, Sean Hefty wrote: > By the time the kernel client gets the MAD, it's been reassembled, and > the exact size that's needed is known. So, I don't think that this is > an issue for kernel clients. > > Maybe you could add a peek for usermode or have read return the correct > size if the requested size is too small.
I especially like the requested size too small touch. That saves a potential extra operation. > Since you have to do a data > copy for usermode anyway, I think it makes sense to just return the > coalesced buffer. This makes me think that I should implement > ib_coalesce_recv_mad() now. That would make the receive side of user space much easier. Shall I defer implementing this part until coalesce is implemented ? Thanks. -- Hal _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
