Bufferbloat

09/11/2011

02:54 pm Cerowrt Bug #262: poor wireless performance in smoketest rc6

My test had neither Babel nor multiple interfaces (but did bridge both radios and the lan together). So, it's not Ba...

09/10/2011

11:52 pm Cerowrt Bug #262: poor wireless performance in smoketest rc6

Important to benchmarking this stuff, although it is not in play in the present test, since Nagle has nothing to do w...

05:26 pm Cerowrt Bug #262: poor wireless performance in smoketest rc6

It's something to do with ICMP rate limiting in OS X. I can reproduce it with my own openwrt build:
root@OpenWrt:~#...

09/06/2011

03:50 am Cerowrt Bug #262: poor wireless performance in smoketest rc6

Ok, if it's that consistent, I'll call the test valid.
Dave, a few millimeters can be significant... but if you get ...

02:25 am Cerowrt Bug #262: poor wireless performance in smoketest rc6

So, I have one issue with the test methodology... equidistant doesn't cut it, you need to test the same pair of devic...

08/05/2011

11:39 am Cerowrt Bug #216: Infiinite retry for ping

Ok, this patch applies on top of Felix' last set of changes, and produces really spectacular results: about 2x better...

08/03/2011

03:44 pm Cerowrt Bug #216: Infiinite retry for ping

Here's what I've been working on. Note that this includes a version of Felix' patch that changes the interpretation ...

02:08 pm Cerowrt Bug #216: Infiinite retry for ping

Just starting to roll them up.
I got g to suck much less.
Note that although I have not reduced ATH_TXBUF in these,...

09:16 am Cerowrt Bug #216: Infiinite retry for ping

Tickless was invented by SGI, and the MIPS CPU architecture was designed to run tickless, so it's nearly optimal for ...

08:16 am Cerowrt Bug #216: Infiinite retry for ping

David Taht wrote:
> In looking over your tests I suspect the reason your latency under load
> remains good is that ...