> On Aug 6, 2015, at 5:39 PM, Jarno Rajahalme <jrajaha...@nicira.com> wrote: > >> On Aug 6, 2015, at 16:29, Jesse Gross <je...@nicira.com> wrote: > >> >>> On Thu, Aug 6, 2015 at 4:17 PM, Jarno Rajahalme <jrajaha...@nicira.com> >>> wrote: >>> Struct miniflow is now sometimes used just as a map. Define a new >>> struct flowmap for that purpose. The flowmap is defined as an array of >>> maps, and it is automatically sized according to the size of struct >>> flow, so it will be easier to maintain in the future. >>> >>> It would have been tempting to use the existing struct bitmap for this >>> purpose. The main reason this is not feasible at the moment is that >>> some flowmap algorithms are simpler when it can be assumed that no >>> struct flow member requires more bits than can fit to a single map >>> unit. The tunnel member already requires more than 32 bits, so the map >>> unit needs to be 64 bits wide. >>> >>> Performance critical algorithms enumerate the flowmap array units >>> explicitly, as it is easier for the compiler to optimize, compared to >>> the normal iterator. Without this optimization a classifier lookup >>> without wildcard masks would be about 25% slower. >>> >>> With this more general (and maintainable) algorithm the classifier >>> lookups are about 5% slower, when the struct flow actually becomes big >>> enough to require a second map. This negates the performance gained >>> in the "Pre-compute stage masks" patch earlier in the series. >>> >>> Requested-by: Ben Pfaff <b...@nicira.com> >>> Signed-off-by: Jarno Rajahalme <jrajaha...@nicira.com> >> >> I think this patch may need a rebase after the tunnel metadata flow >> patches that I applied last night. >> >> Sorry about that. > > Thanks for the heads up! I just sent a v3. >
I did some more testing and profiling and sent an improved version (v4). No changes in the first few patches. Jarno _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
