[rt2x00-users] [bugzilla-daemon at bugzilla.kernel.org: [Bug 42828] rt2800pci unstable - chokes after too much I/O]

Francisco Pina Martins f.pinamartins at gmail.com
Mon Nov 5 08:54:05 AEDT 2012


Hello again,

Sorry I have been quiet for so long. I have been away from home again and
did not have the chance to retry my testing.

Anyway, great work on those graphs! I can see how radically different they
are from mine, and in fact you have much better throughput.
However, I think I have an explanation for it. The machine I have been
using as my netperf server is a rapsberry pi. I have now realized that the
speed of the ethernet port is not optimal at all. I will retest with a
different server now and report back.
If the results are still "strange" I will dig a bit in power management.
I should report back soon.

Francisco

On 4 November 2012 15:20, Andreas Hartmann <andihartmann at 01019freenet.de>wrote:

> On Fri, Oct 26, 2012 21:28:17 +0100
> Francisco Pina Martins <f.pinamartins at gmail.com> wrote:
>
> [...]
>
> > On 26 October 2012 16:14, Andreas Hartmann
> > <andihartmann at 01019freenet.de>wrote:
> >
> > > Hello Francisco,
> > >
> > > Francisco Pina Martins wrote:
> > > > Ok, so here are the results of my testing.
> > > > I used Andreas' script for netperf under 2 different receiving
> > > > conditions (good and poor) and under 3 different kernels (lts
> > > > a.k.a. control
> > > (3.0.45),
> > > > bugged (3.5.0), patched (3.5.0 with Gertjan's patch)).
>
> Did you already try to disable powermanagement? Enabled powermanagement
> can produce strange behavior as I learned today with rt2800usb. But I
> don't know, if the pci driver enables powermanagement per default, too.
>
> What gives
>
> iwlist wlan0 power
>
> If it is enabled (Current mode: on), you could try to disable it with
>
> iwconfig wlan0 power off
>
> and test again :-).
>
>
> Kind regards,
> Andreas
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rt2x00.serialmonkey.com/pipermail/users_rt2x00.serialmonkey.com/attachments/20121104/1ca11cfe/attachment-0002.html>


More information about the users mailing list