Re: [B.A.T.M.A.N.] [RFC 15/23] batman-adv: ELP - send unicast ELP packets for throughput sampling

[Antonio Quartulli <antonio@meshcoding.com>]

[-- Attachment #1: Type: text/plain, Size: 2435 bytes --]

On 12/02/14 13:56, Antonio Quartulli wrote:
> On 12/02/14 13:54, Felix Fietkau wrote:
>> On 2014-02-12 13:12, Antonio Quartulli wrote:
>>> On 12/02/14 10:12, Andrew Lunn wrote:
>>>> On Tue, Feb 11, 2014 at 01:48:15PM +0100, Antonio Quartulli wrote:
>>>>> From: Antonio Quartulli <antonio@open-mesh.com>
>>>>>
>>>>> In case of a unused link, the throughput estimation will
>>>>> get stuck to the last sampled value and therefore the
>>>>> reported metric will becomes obsolete.
>>>>>
>>>>> Send unicast ELP packets to each neighbor to trigger throughput
>>>>> sampling on unused links.
>>>>
>>>> Humm, i can understand the need for this, but i really think the rate
>>>> control code should be sending the probes, not batman. What do the
>>>> wifi people say about this? Have they tried submitting patches to
>>>> them?
>>>
>>> I am CC'ing Felix so that he can give his opinion.
>>> But last time I checked Minstrel I realised that it uses data packets to
>>> probe rates (there is not a specific probing packet), meaning that if
>>> there is no data packet to send, then no probing is performed.
>> Correct. Minstrel never sends dedicated probing packets. Changing
>> minstrel to send active probing packets to serve the needs of batman
>> would be a very bad idea, because pretty much all regular users do not
>> have any need for extra probing.
>>
>>> Sending this ELP packets (when there is no unicast traffic) is a way to
>>> trigger this mechanism in Minstrel.
>> Right. If I remember correctly, if you send less than 1 packet per 100
>> ms, then all packets should end up being probing packets. If that
>> doesn't work, we can change minstrel's probing pattern to allow things
>> like batman to get a desirable amount of rate control probing simply by
>> sending unicast packets.
> 
> This is what I am trying to achieve here: if no unicast packets have
> been sent for the last 100ms (at least) then send N probing packets in a
> row (with N = 2 at the moment - it is a define in main.h).
> 

However, from the experiments performed so far it looked like Minstrel
was behaving properly under these batman probing packets.


As soon as we move on with this project I am sure we will get some more
feedback from other users (perhaps this is something we can test at the
WBM? - this code should be ready and usable by then - cross fingers)


Cheers,


-- 
Antonio Quartulli


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]