RE： weird powersave behavior

[claud_yu]

Hi Holger , 

That is normal. The comand "iwconfig eth1 power on" will enable the power management facility of wifi chip. 
Accroding 802.11 spec , that define a power-saving mode (PSM) to reduce the power consumption. The mode turn off the TX/RX periodically (every super-frame). 
Every STA need to wake during every beacon to maintain connection state and wireless traffic. But that will introduce latency ( multiple of beacon interval). 
Every wifi chip with PSM have the same behavior.


Claud Yu



--- 08/4/29 (星期二)，Holger Schurig <hs4233 at mail.mn-solutions.de> 寫道：

> 寄件者: Holger Schurig <hs4233 at mail.mn-solutions.de>
> 主旨: weird powersave behavior
> 收件者: libertas-dev at lists.infradead.org
> 日期: 2008 4 29 星期二 下午 7:30
> Can anyone explain this behavior?
> 
> When I do "iwconfig eth1 power on" on my target
> and ping the 
> target from another host, I get "ramps" on the
> response times:
> 
> 
> 
> with "iwconfig eth1 power off" or the default
> settings:
> 
> 64 bytes from 10.2.1.4: icmp_seq=1 ttl=64 time=2.59 ms
> 64 bytes from 10.2.1.4: icmp_seq=2 ttl=64 time=1.67 ms
> 64 bytes from 10.2.1.4: icmp_seq=3 ttl=64 time=1.65 ms
> 64 bytes from 10.2.1.4: icmp_seq=4 ttl=64 time=1.66 ms
> 64 bytes from 10.2.1.4: icmp_seq=5 ttl=64 time=1.95 ms
> 64 bytes from 10.2.1.4: icmp_seq=6 ttl=64 time=1.68 ms
> 64 bytes from 10.2.1.4: icmp_seq=7 ttl=64 time=1.96 ms
> 64 bytes from 10.2.1.4: icmp_seq=8 ttl=64 time=1.66 ms
> 64 bytes from 10.2.1.4: icmp_seq=9 ttl=64 time=1.65 ms
> 
> with "iwconfig eth1 power on":
> 
> 64 bytes from 10.2.1.4: icmp_seq=11 ttl=64 time=101 ms
> 64 bytes from 10.2.1.4: icmp_seq=12 ttl=64 time=121 ms
> 64 bytes from 10.2.1.4: icmp_seq=13 ttl=64 time=141 ms
> 64 bytes from 10.2.1.4: icmp_seq=14 ttl=64 time=161 ms
> 64 bytes from 10.2.1.4: icmp_seq=15 ttl=64 time=181 ms
> 64 bytes from 10.2.1.4: icmp_seq=16 ttl=64 time=201 ms
> 64 bytes from 10.2.1.4: icmp_seq=17 ttl=64 time=16.1 ms
> 64 bytes from 10.2.1.4: icmp_seq=18 ttl=64 time=36.5 ms
> 64 bytes from 10.2.1.4: icmp_seq=19 ttl=64 time=56.4 ms
> 64 bytes from 10.2.1.4: icmp_seq=20 ttl=64 time=76.4 ms
> 64 bytes from 10.2.1.4: icmp_seq=21 ttl=64 time=96.4 ms
> 64 bytes from 10.2.1.4: icmp_seq=22 ttl=64 time=116 ms
> 64 bytes from 10.2.1.4: icmp_seq=23 ttl=64 time=136 ms
> 64 bytes from 10.2.1.4: icmp_seq=24 ttl=64 time=157 ms
> 64 bytes from 10.2.1.4: icmp_seq=25 ttl=64 time=177 ms
> 64 bytes from 10.2.1.4: icmp_seq=26 ttl=64 time=196 ms
> 64 bytes from 10.2.1.4: icmp_seq=27 ttl=64 time=10.9 ms
> 64 bytes from 10.2.1.4: icmp_seq=28 ttl=64 time=30.8 ms
> 64 bytes from 10.2.1.4: icmp_seq=29 ttl=64 time=50.9 ms
> 64 bytes from 10.2.1.4: icmp_seq=30 ttl=64 time=71.2 ms
> 64 bytes from 10.2.1.4: icmp_seq=31 ttl=64 time=91.0 ms
> 64 bytes from 10.2.1.4: icmp_seq=32 ttl=64 time=112 ms
> 
> Every ping takes 20ms more than the previous ...
> 
> _______________________________________________
> libertas-dev mailing list
> libertas-dev at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/libertas-dev


      _________________________________________________________________________________________________________
蚊子吵得整晚睡不著，一直叮我，誰能救救我？ http://ksk.search.yahoo.net/index.asp?WT.mc_id=023