I was actually thinking that when the limit is reached, we would stop creating new batches, and simply execute any packets which don't fit into the existing set one by one. Not sure which way is better. Anyways I'll send out a proposal and we can decide how to do it in the review.
Ethan On Mon, Jun 30, 2014 at 2:51 PM, Jarno Rajahalme <jrajaha...@nicira.com> wrote: > > On Jun 30, 2014, at 2:32 PM, Ethan Jackson <et...@nicira.com> wrote: > >>> Would it make sense to batch the upcalls as well, i.e. collect them up and >>> then issue them with a single dp_netdev_output_userspace()? >> >> Yep, Ryan is working on that. >> >>> It would seem that starting from the newest batch and looping backwards >>> might find the batch sooner, especially if the packets of a batch arrive >>> back to back. >> >> True. I talked about this offline with Pravin, and we think that the >> best thing is to simply limit the number of packet_batches you're >> allowed to have per batch. If it's something like 4 then it shouldn't >> matter. Does that sound reasonable? If so I'll do it as a separate >> patch. >> > > You mean that when the limit is reached, the collected batches are executed > and then start over with no batches to be executed (n_batches = 0)? Sounds OK > for a separate batch. > > Jarno > _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
