> > > I can't get more than 10 or 11 megabits throughput between a pair of XOs > sitting right next to eachother. That's XO-XO via the mesh. The XO says the > links are running at 54 megabits so I'm expecting a lot more than that. >
54 mbps is the bit rate without consideration of the time it takes to get onto the air and the various framing overheads. I did a lot of work to make the NANDblaster send as fast as possible, using the mesh in a restricted mode to avoid retransmissions. I got to 2.8 Mbytes/sec. Ricardo Carrano told me that the fastest rate that has ever been measured using Linux is 1.4 MBytes/sec. I can easily explain the 1.4 MBytes/sec figure - that's exactly what you'll get if you send maximum-size Ethernet frames at a rate of one frame per millisecond. The 1/ms rate limitation is easily explained by the USB macro-frame time. To get to the 2.8 rate, I had to fiddle with OFW's USB stack to make sure that I was queuing up packets just right so the USB host controller could send more than one without going through the "service other schedules in this macroframe" dance. I did some tests where I pushed data as fast as possible to the chip, but configured the chip not to actually transmit them. The rate under Linux was the aformentioned 1.4 MB/sec, confirming the USB + driver bottleneck hypothesis. Under OFW, I got up to 4.2 MB/sec. > > I see occasional glitches that reduce throughput. I think that's somebody > taking time out to go probe for APs and such and the radio is deaf when > working on another channel. > For the nandblaster, I went to some effort to ensure that the XOs weren't contributing to the beacon traffic. I had to turn of several things. It was complicated. > Where is that code? Is that all at the Sugar level or does the low level > WiFi/NetworkManager stuff do similar nasty things? How often should I expect > disruptions? (I'm seeing them every minute or two I haven't looked > carefully.) > Some of it is handled inside the wireless module firmware. Perhaps not all. _______________________________________________ Devel mailing list [email protected] http://lists.laptop.org/listinfo/devel
