Development issues regarding the cerowrt test router project
 help / color / mirror / Atom feed
* [Cerowrt-devel] some ath9k rework in progress
@ 2013-07-21 21:13 Dave Taht
  2013-07-21 21:19 ` Felix Fietkau
  0 siblings, 1 reply; 3+ messages in thread
From: Dave Taht @ 2013-07-21 21:13 UTC (permalink / raw)
  To: cerowrt-devel, Felix Fietkau

https://github.com/mirrors/openwrt/commit/0362bf8e3d54a6ab637fee9746e3fd3ec9a5db57
ath9k: add initial tx queueing rework patches

This forces all packets (even for un-aggregated traffic) through
software queues to improve fairness and stability

Signed-off-by: Felix Fietkau

Although felix and I discussed this a few months back, (I have a
transcription of the convo I should clean up and post) I'm not sure
where this lies on the road ahead....

-- 
Dave Täht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [Cerowrt-devel] some ath9k rework in progress
  2013-07-21 21:13 [Cerowrt-devel] some ath9k rework in progress Dave Taht
@ 2013-07-21 21:19 ` Felix Fietkau
  2013-07-21 21:25   ` Dave Taht
  0 siblings, 1 reply; 3+ messages in thread
From: Felix Fietkau @ 2013-07-21 21:19 UTC (permalink / raw)
  To: Dave Taht; +Cc: cerowrt-devel

On 2013-07-21 11:13 PM, Dave Taht wrote:
> https://github.com/mirrors/openwrt/commit/0362bf8e3d54a6ab637fee9746e3fd3ec9a5db57
> ath9k: add initial tx queueing rework patches
> 
> This forces all packets (even for un-aggregated traffic) through
> software queues to improve fairness and stability
> 
> Signed-off-by: Felix Fietkau
> 
> Although felix and I discussed this a few months back, (I have a
> transcription of the convo I should clean up and post) I'm not sure
> where this lies on the road ahead....
This doesn't do active queue management yet, but it does help with
preventing legacy traffic from flooding the queues in a way that hurts
aggregated traffic. It also helps mitigate the impact of multicast
flooding, or flooding one particular station with traffic.

Next step ahead is moving that queueing to mac80211, I'm going to be
working on that soon.

- Felix


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [Cerowrt-devel] some ath9k rework in progress
  2013-07-21 21:19 ` Felix Fietkau
@ 2013-07-21 21:25   ` Dave Taht
  0 siblings, 0 replies; 3+ messages in thread
From: Dave Taht @ 2013-07-21 21:25 UTC (permalink / raw)
  To: Felix Fietkau; +Cc: cerowrt-devel

On Sun, Jul 21, 2013 at 2:19 PM, Felix Fietkau <nbd@openwrt.org> wrote:
> On 2013-07-21 11:13 PM, Dave Taht wrote:
>> https://github.com/mirrors/openwrt/commit/0362bf8e3d54a6ab637fee9746e3fd3ec9a5db57
>> ath9k: add initial tx queueing rework patches
>>
>> This forces all packets (even for un-aggregated traffic) through
>> software queues to improve fairness and stability
>>
>> Signed-off-by: Felix Fietkau
>>
>> Although felix and I discussed this a few months back, (I have a
>> transcription of the convo I should clean up and post) I'm not sure
>> where this lies on the road ahead....
> This doesn't do active queue management yet, but it does help with
> preventing legacy traffic from flooding the queues in a way that hurts
> aggregated traffic. It also helps mitigate the impact of multicast
> flooding, or flooding one particular station with traffic.
>
> Next step ahead is moving that queueing to mac80211, I'm going to be
> working on that soon.

Thanks for the update! I'm mostly busying myself building a lab that
can actually exercise this stuff fully...

https://plus.google.com/s/%23yurtlab

> - Felix
>



-- 
Dave Täht

Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.html

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2013-07-21 21:25 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-07-21 21:13 [Cerowrt-devel] some ath9k rework in progress Dave Taht
2013-07-21 21:19 ` Felix Fietkau
2013-07-21 21:25   ` Dave Taht

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox