Re: [RFC PATCH 1/2] mac80211: Check for queued frames before entering power save.

[Vivek Natarajan <vivek.natraj@gmail.com>]

On Thu, Apr 7, 2011 at 5:21 PM, Johannes Berg <johannes@sipsolutions.net> wrote:
> On Thu, 2011-04-07 at 17:17 +0530, Vivek Natarajan wrote:
>
>> > I'm not sure I understand. Where does the 100ms come from? This code
>> > flushes the TX queues, which can take as much time as it wants, no? How
>> > does it break applications?
>>
>> 100ms is the time after which the dynamic ps timer stops the netif
>> queues and flushes all the frames in the driver. Actually two factors
>> caused the application to timeout:
>>  One is the delay in the flush() for which the netif queues are
>> stopped and eventually the application could not send out the frames.
>> The second is the dropping of frames in the flush() callback when it
>> reaches the timeout period.
>
> Is that an ath9k specific thing? mac80211 never invokes flush() with
> drop=true, at this point. I thought we needed it and added it, but I
> guess we don't really need it.

Currently, ath9k gives a timeout of 200 ms for the pending frames to
complete. After this timeout, the pending frames will be dropped even
if drop is set as false in mac80211.

Vivek.