Michael Buesch wrote:
> On Thursday 02 August 2007, Larry Finger wrote:
>> Michael,
>>
>> I couldn't find any long/short indication in the header, so I added a bool 
>> that
>> is set when the frame is sent.
> 
> This is not going to work.
> 
> But we can do this differently.
> You can do something like:
> 
> if (!status->acked && !tx_control->noack)
>       excessive_retries = 1;
> 
> So we don't need to care about the retry count.
> 
> Anyway. I don't know why we need excessive_retries _at_ _all_.
> The rc algorithm does already know if the frame succeed or failed
> anyway.

Yes, it does, but it only looks at the potential for reduction if excessive 
retries is set. Why? 
Because it was coded that way! As I said earlier, I don't want to get involved 
there.

I'll try to implement it as above.

Larry

_______________________________________________
Bcm43xx-dev mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

Reply via email to