Re: [PATCH bluetooth-next 2/3] ieee802154/atusb: Mark driver as AACK enabled in hardware.

[Alexander Aring <alex.aring@gmail.com>]

On Thu, May 28, 2015 at 01:36:14PM +0300, Lennert Buytenhek wrote:
...
> > 
> > But one of these modes (max_frame_retries = -1 OR max_frame_retries >=
> > 0) should be supported and AACK handling should also always supported.
> 
> Hardware that only supports max_frame_retries = -1 is clearly not
> 802.15.4 compliant -- how much of such hardware is there out there?
> 

This is correct, because 802.15.4 default is max_frame_retries = 3. I
never saw a transceiver which doesn't support ARET handling in hardware
by transceiver.

What we do currently is:

If a transceiver doesn't support handling of setting max_frame_retries
parameters then we assume the 802.15.4 defaults. We can't say if it's
really supported or not, depends on driver implementation.

And currently for max_frame_retries we assume "-1" which is not correct,
because 802.15.4 describes "3" as default. But again, this is only to
hold backwardscompability with nodes which doesn't support AACK.

> 
> > Maybe I should draw some graphics to display what the meaning of
> > max_frame_retries really is. At the end each driver should follow all
> > the same meaning of this value.
> 
> I think it's clear to me.

ok.

- Alex