2008-12-02 19:27:36

by Andrey Yurovsky

[permalink] [raw]
Subject: Re: Problems with o11s

Hi Keir. I believe you're using ath5k? You can use, at this time, up
to (but not including) commit
dc300b5c5291d779a53d83af960fdf9388bf0544, where we have a regression
in the driver when the PCU-related code was updated. Have you tried
testing IBSS mode in your setup?

-Andrey

On Tue, Dec 2, 2008 at 7:52 AM, Keir <[email protected]> wrote:
> Hi,
>
> I've had o11s working for a while now, but I seem to get a strange problem
> whereby pinging between MPs works for the first few pings, and then I start
> getting packet loss (Destination Host Unreachable) and nothing gets
> through, despite the other station being listed as "ESTAB" by iw. I'm not
> sure if there is something strange about my setup, or this is a bug
> somewhere, anyone got any ideas? I'm using revision
> d059b2aed9e93cb94364e01374bca90ce3dc0ef9
>
> Cheers
>
> Keir
>


2008-12-08 15:35:18

by Keir

[permalink] [raw]
Subject: Re: Problems with o11s

Hi Andrey,

Using the commit directly before
dc300b5c5291d779a53d83af960fdf9388bf0544 (I'm pretty sure). Don't
know how to test IBSS. I'm guessing the problem here is the
plink_close messages arent getting through somehow?

Cheers for all your help

Keir

2008/12/2 Andrey Yurovsky <[email protected]>:
> Hi Keir. I believe you're using ath5k? You can use, at this time, up
> to (but not including) commit
> dc300b5c5291d779a53d83af960fdf9388bf0544, where we have a regression
> in the driver when the PCU-related code was updated. Have you tried
> testing IBSS mode in your setup?
>
> -Andrey
>
> On Tue, Dec 2, 2008 at 7:52 AM, Keir <[email protected]> wrote:
>> Hi,
>>
>> I've had o11s working for a while now, but I seem to get a strange problem
>> whereby pinging between MPs works for the first few pings, and then I start
>> getting packet loss (Destination Host Unreachable) and nothing gets
>> through, despite the other station being listed as "ESTAB" by iw. I'm not
>> sure if there is something strange about my setup, or this is a bug
>> somewhere, anyone got any ideas? I'm using revision
>> d059b2aed9e93cb94364e01374bca90ce3dc0ef9
>>
>> Cheers
>>
>> Keir
>>
>