On Fri, Dec 2, 2011 at 8:01 AM, Fabrice Deyber <[email protected]> wrote: > Hi all, > sorry for the delay, I was distracted by other tasks. > So, I had a chance to run a multi-hop test on a secured mesh. > I used up to 4 nodes for the hop test. Again I'm not using any HT mode. > I ran Jperf from PCs connected to the nodes via ethernet (nodes are > setup in bridge mode, i.e. mesh and ethernet interface are bridged > together). I ran the TCP tests with a 16KB buffer size on channel 149. > > Node 1 -> Node2 (1 hop): ~17Mbps > Node 1 -> Node 2 -> Node 3 (2 hops): ~8.5 Mbps > Node 1 -> Node 2 -> Node 3 -> Node 4 (3 hops): ~4Mbps > > The results were similar on both directions.
Great info, thanks! > Note : I forced the path using iw mesh0 set mpath ... next_hop ... > (I noticed that forcing mesh_max_peer_links to 1 does not work) Does not work at all (i.e. you get more than 1 peer link)? Or you don't get the linear topology that you were expecting? Cheers, Javier _______________________________________________ Devel mailing list [email protected] http://open80211s.com/mailman/listinfo/devel
