Re: [RFC] mac80211: budget outstanding airtime for transmission

2018-09-21 Thread Kalle Valo
Rajkumar Manoharan writes: >>> diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h >>> index b1b0fd6a2e21..ddc2c882c91c 100644 >>> --- a/net/mac80211/sta_info.h >>> +++ b/net/mac80211/sta_info.h >>> @@ -135,6 +135,7 @@ struct airtime_info { >>> u64 rx_airtime; >>> u64

Re: [RFC] mac80211: budget outstanding airtime for transmission

2018-09-20 Thread Dave Taht
As a side note (good work!) - I would dearly like to visibly account for management frames somewhere that can be seen from userspace. ?

Re: [RFC] mac80211: budget outstanding airtime for transmission

2018-09-20 Thread Rajkumar Manoharan
On 2018-09-20 01:30, Toke Høiland-Jørgensen wrote: Rajkumar Manoharan writes: Per frame airtime estimation could be used to track outstanding airtime of each txq and can be used to throttle ieee80211_tx_dequeue(). This mechanism on its own will get us the queue limiting and latency reduction

Re: [RFC] mac80211: budget outstanding airtime for transmission

2018-09-20 Thread Toke Høiland-Jørgensen
Rajkumar Manoharan writes: > Per frame airtime estimation could be used to track outstanding airtime > of each txq and can be used to throttle ieee80211_tx_dequeue(). This > mechanism on its own will get us the queue limiting and latency > reduction goodness for firmwares with deep queues. And

[RFC] mac80211: budget outstanding airtime for transmission

2018-09-20 Thread Rajkumar Manoharan
Per frame airtime estimation could be used to track outstanding airtime of each txq and can be used to throttle ieee80211_tx_dequeue(). This mechanism on its own will get us the queue limiting and latency reduction goodness for firmwares with deep queues. And for that it can be completely