On Sat, 2010-06-12 at 19:47 +0200, Johannes Berg wrote:
> On Sat, 2010-06-12 at 13:34 -0400, Pavel Roskin wrote:
> > On Fri, 2010-06-11 at 10:27 -0700, Jouni Malinen wrote:
> >
> > > - for (i = 0; i < NUM_RX_DATA_QUEUES; i++) {
> > > + for (i = 0; i < NUM_RX_DATA_QUEUES + 1; i++) {
> >
> > Perhaps we could have a define for NUM_RX_DATA_QUEUES + 1, e.g.
> > NUM_RX_ALL_QUEUES
>
> I kinda disagree. Yes, this is tricky code, but adding a define wouldn't
> make it clearer. In fact, it isn't really related to queues to start
> with, and ALL_QUEUES would just strengthen that mostly wrong notion.
That's certainly not my intention to strengthen wrong notions.
> I guess it really should be renamed to TIDs with the last two being
> special for non-QoS (so no TID) traffic + mgmt traffic. But then
> sequence numbers are allocated from the same counter (so there we just
> have 17 possibilities) while PNs have 18 counters...
Then let's replace QUEUE with TID in all RX defines. Two special TIDs
could be used for non-QoS traffic and management frames when
appropriate.
--
Regards,
Pavel Roskin
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
