[PATCH 0/3] RFC: fix bug http://docs.openmoko.org/trac/ticket/1884

Werner Almesberger werner at openmoko.org
Sun Nov 16 21:28:30 CET 2008


I wrote:
> One oddity: when I ran my performance tests to see how much of a
> difference single-bit mode made (almost disappointingly, it's only
> a bit slower, ~645kB/s host to neo, and ~665kB/s neo to host), the
> receiver hung in one test run. That's the first time I ever saw
> such a problem.
> 
> Wonder if this was just a freak accident ... well, I have a little
> torture test set up for tonight that will tell.

In 500 transfers host->neo, 499 were successful while one failed
because the receiver wasn't ready. This seems to be a flaw in my
test setup then, and not a WLAN problem. No problems in the
opposite direction.

In the original test that hung (i.e., the one because of which I
did the torture test), it seems that the Neo temporarily lost
association:

(good)  ttcp-t: 16777216 bytes in 25.83 real seconds = 634.35 KB/sec +++
(bad)   ttcp-t: 9699328 bytes in 1055.29 real seconds = 8.98 KB/sec +++
(after) ttcp-t: connect: No route to host
(after) ttcp-t: connect: No route to host
...

So, no smoking gun here. Maybe Murphy was just pulling my leg.

- Werner



More information about the openmoko-kernel mailing list