AR6000 netif_queue_stop non stop, Bug?
ivan_p at hotbox.ru
Wed Apr 1 03:03:13 CEST 2009
Changes: Data queue in driver reducted to one packet, for use kernel packet management engine.
Changes: Queue WMI_CONTROL_PRI full/available control fully moved to ar6000_tx_queue_full/ar6000_tx_queue_available.
Changes: Small cleanup.
Additional advantages at reducted data queue.
At this moment driver work in pseudo asynchronous mode, it mean, what transfer start only after 'io' thread take control, before it
moment all packets will collected in hif2 layer queue. Critical packet - packet with frlag on request additional credits. If we have
prancing (saltatory) trafic from kernel side, this packet will send only after driver queue will fully filled or kernel queue will
empty. In this case we get conveyor fail, and lost of time (about ms). Reducting queue length allow to fast send the critical
packcket and receive back new credits.
> About this... After work, I will prepare patch (from my experemental src), for testing queue reducted to 1 item. I'am already sent
> it to Dongas for testing few days ago,
> And try comment additional ussage advantages on hif2 (present to much fast reaction).
For werner, sorry today not read Documents\..., if not fully cleanup, please write, I will correct it.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 2307 bytes
Desc: not available
Url : http://lists.openmoko.org/pipermail/openmoko-kernel/attachments/20090401/b7132aa5/attachment.obj
More information about the openmoko-kernel