[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[oiioeu at aol.com]

 

 I got avg. 60Mbps of the throughput rate by iperf testing and had the testing environment like below:


OpenWrt:                       r18529 (I tried the latest r20851  and the throughput rate could be even worse )
AP && STA h/w platform: IXP425 
wireless module:             AR9280( 11na, HT40 )
wireless driver:                ath9k version matched OpenWrt r18529
topology is like:               PC1 - STA - AP - PC2 ( ethernet connection between PC1 and STA, AP and PC2, and the wireless 
                                      connection between STA and the AP. iperf client on PC1 and iperf server on PC2 )

thenI added oprofile support to this throughput rate testing and had thethroughput rate decrease to avg 8Mbps, here are my questions:
1. whether the avg. 8Mbps of the throughput after adding the oprofile support seemed to be reasonable?
2. why there were the warning messeges at the beginning of the opreport result and does it matter or not?
3. anyone could point me is there anything abnormal from the opreport messeges that lead to the low throughput in my case?
the oprofile log are attached 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100419/52c788ec/attachment-0001.htm 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: oprofile_AP.log
Url: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100419/52c788ec/attachment-0002.txt 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: oprofile_SU.log
Url: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100419/52c788ec/attachment-0003.txt 

[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[Dave Taht]

On 04/19/2010 07:05 PM, oiioeu at aol.com wrote:
>
> I got avg. 60Mbps of the throughput rate by iperf testing and had the 
> testing environment like below:
>
>
> OpenWrt:                       r18529 (I tried the latest r20851  and 
> the throughput rate could be even worse )
> AP && STA h/w platform: IXP425
> wireless module:             AR9280( 11na, HT40 )
> wireless driver: ath9k version matched OpenWrt r18529
> topology is like:               PC1 - STA - AP - PC2 ( ethernet 
> connection between PC1 and STA, AP and PC2, and the wireless
>                                       connection between STA and the 
> AP. iperf client on PC1 and iperf server on PC2 )
>
> then I added oprofile support to this throughput rate testing and had 
> the throughput rate decrease to avg 8Mbps, here are my questions:
> 1. whether the avg. 8Mbps of the throughput after adding the oprofile 
> support seemed to be reasonable?
> 2. why there were the warning messeges at the beginning of the 
> opreport result and does it matter or not?
> 3. anyone could point me is there anything abnormal from the opreport 
> messeges that lead to the low throughput in my case?
> the oprofile log are attached

Interesting. I've been meaning to get around to oprofiling my hardware 
(nanostation M5) one of these days  as I'd noted that I was achieving 
~70Mbit via iperf and seeing (via top) that the hardware was spending 
70% of its time in the "sirq" state. (whether or not that was the 
wireless or wired drivers doing that, I don't know).

oprofile tends to be high overhead unless there is special on-chip 
counter support.

As for this particular oprofile, I don't know, but it looks like 
ath9k_ioread32 is a good candidate for optimization, even without the 
huge oprofile slowdown....


>
> _______________________________________________
> ath9k-devel mailing list
> ath9k-devel at lists.ath9k.org
> https://lists.ath9k.org/mailman/listinfo/ath9k-devel
>    

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100428/cd7523f8/attachment.htm 

[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[Luis R. Rodriguez]

On Wed, Apr 28, 2010 at 07:17:47AM -0700, Dave Taht wrote:
> On 04/19/2010 07:05 PM, oiioeu at aol.com<mailto:oiioeu@aol.com> wrote:
> 
> I got avg. 60Mbps of the throughput rate by iperf testing and had the testing environment like below:
> 
> 
> OpenWrt:                       r18529 (I tried the latest r20851  and the throughput rate could be even worse )
> AP && STA h/w platform: IXP425
> wireless module:             AR9280( 11na, HT40 )
> wireless driver:                ath9k version matched OpenWrt r18529
> topology is like:               PC1 - STA - AP - PC2 ( ethernet connection between PC1 and STA, AP and PC2, and the wireless
>                                       connection between STA and the AP. iperf client on PC1 and iperf server on PC2 )
> 
> then I added oprofile support to this throughput rate testing and had the throughput rate decrease to avg 8Mbps, here are my questions:
> 1. whether the avg. 8Mbps of the throughput after adding the oprofile support seemed to be reasonable?
> 2. why there were the warning messeges at the beginning of the opreport result and does it matter or not?
> 3. anyone could point me is there anything abnormal from the opreport messeges that lead to the low throughput in my case?
> the oprofile log are attached
> 
> Interesting. I've been meaning to get around to oprofiling my hardware (nanostation M5) one of these days  as I'd noted that I was achieving ~70Mbit via iperf and seeing (via top) that the hardware was spending 70% of its time in the "sirq" state. (whether or not that was the wireless or wired drivers doing that, I don't know).
> 
> oprofile tends to be high overhead unless there is special on-chip counter support.
> 
> As for this particular oprofile, I don't know, but it looks like ath9k_ioread32 is a good candidate for optimization, even without the huge oprofile slowdown....

ath9k_ioread32() could be made simpler by becoming a callback after
we check for the serialization requirement. We can use a default
callback (the current one) for pre-initialization and then then we
can just peg the proper callback upon ath9k_hw_attach_ops(). The
callback for the case where serialization is not required would be
a straight forward read.

This can be done for ath9k_iowrite32() as well.

  Luis

[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[oiioeu at aol.com]

 

 thank you Luis and Dave for the answers about distance issue, I'd like to ask again why CPU is idle for 30%~ 40% while the throughput testing only gave an avg. 50Mbps. any mean I could improve that? I've been experiencing low throughput on IXP4XX platform since last year, please.


 

 

-----Original Message-----
From: Luis R. Rodriguez <lrodriguez@atheros.com>
To: oiioeu <oiioeu@aol.com>
Cc: Luis Rodriguez <Luis.Rodriguez@Atheros.com>; d <d@teklibre.org>; ath9k-devel <ath9k-devel@lists.ath9k.org>
Sent: Tue, May 25, 2010 11:51 pm
Subject: Re: [ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions


On Tue, May 25, 2010 at 01:24:01AM -0700, oiioeu at aol.com wrote:
> I found:
> I got an average throughput about 6~7Mbps because I put the AP and STA very 
close(side by side), today I just put them 10 meters away from each other and I 
got an avg. throughput rate like 50Mbps, with latest OpenWrt r21560. I don't 
understant why put the AP and STA side by side would come out very low 
throughput, anyway. I did the oprofiling again by iperf for 3 times, and the log 
is attached.

For 802.11n it helps if you create different signals on each antenna, so 
distance
is usually a good way to do this. It allows for the objects in the way to take
part of the diversification of the signals. If you are not using an AP with
802.11n then forget about what I just said.

  Luis


 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100525/14469214/attachment.htm 

[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[Luis R. Rodriguez]

On Tue, May 25, 2010 at 01:24:01AM -0700, oiioeu at aol.com wrote:
> I found:
> I got an average throughput about 6~7Mbps because I put the AP and STA very close(side by side), today I just put them 10 meters away from each other and I got an avg. throughput rate like 50Mbps, with latest OpenWrt r21560. I don't understant why put the AP and STA side by side would come out very low throughput, anyway. I did the oprofiling again by iperf for 3 times, and the log is attached.

For 802.11n it helps if you create different signals on each antenna, so distance
is usually a good way to do this. It allows for the objects in the way to take
part of the diversification of the signals. If you are not using an AP with
802.11n then forget about what I just said.

  Luis

[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[Peter Stuge]

oiioeu at aol.com wrote:
> I don't understant why put the AP and STA side by side would come
> out very low throughput, anyway.

The receiver is probably getting saturated, the transmitted signal is
too strong.


//Peter

[ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions

[oiioeu at aol.com]

 I found:

 I got an average throughput about 6~7Mbps because I put the AP and STA very close(side by side), today I just put them 10 meters away from each other and I got an avg. throughput rate like 50Mbps, with latest OpenWrt r21560. I don't understant why put the AP and STA side by side would come out very low throughput, anyway. I did the oprofiling again by iperf for 3 times, and the log is attached. 

avg. 50Mbps of the throughput is still not alright, and I have another question that why the default_idle takes 30~40% of the CPU time? which means the CPU is idle for 30%~40% while the throughput rate is not as good as it could be? what would be the problem?

I'm sorry to be very late, please please answer me, thank you!




 

-----Original Message-----
From: Luis R. Rodriguez <lrodriguez@atheros.com>
To: Dave Taht <d@teklibre.org>
Cc: ath9k-devel <ath9k-devel@lists.ath9k.org>
Sent: Thu, Apr 29, 2010 12:49 am
Subject: Re: [ath9k-devel] oprofile the ath9k for low throughput rate and 3 questions


On Wed, Apr 28, 2010 at 07:17:47AM -0700, Dave Taht wrote:
> On 04/19/2010 07:05 PM, oiioeu at aol.com<mailto:oiioeu@aol.com> wrote:
> 
> I got avg. 60Mbps of the throughput rate by iperf testing and had the testing 
environment like below:
> 
> 
> OpenWrt:                       r18529 (I tried the latest r20851  and the 
throughput rate could be even worse )
> AP && STA h/w platform: IXP425
> wireless module:             AR9280( 11na, HT40 )
> wireless driver:                ath9k version matched OpenWrt r18529
> topology is like:               PC1 - STA - AP - PC2 ( ethernet connection 
between PC1 and STA, AP and PC2, and the wireless
>                                       connection between STA and the AP. iperf 
client on PC1 and iperf server on PC2 )
> 
> then I added oprofile support to this throughput rate testing and had the 
throughput rate decrease to avg 8Mbps, here are my questions:
> 1. whether the avg. 8Mbps of the throughput after adding the oprofile support 
seemed to be reasonable?
> 2. why there were the warning messeges at the beginning of the opreport result 
and does it matter or not?
> 3. anyone could point me is there anything abnormal from the opreport messeges 
that lead to the low throughput in my case?
> the oprofile log are attached
> 
> Interesting. I've been meaning to get around to oprofiling my hardware 
(nanostation M5) one of these days  as I'd noted that I was achieving ~70Mbit 
via iperf and seeing (via top) that the hardware was spending 70% of its time in 
the "sirq" state. (whether or not that was the wireless or wired drivers doing 
that, I don't know).
> 
> oprofile tends to be high overhead unless there is special on-chip counter 
support.
> 
> As for this particular oprofile, I don't know, but it looks like 
ath9k_ioread32 is a good candidate for optimization, even without the huge 
oprofile slowdown....

ath9k_ioread32() could be made simpler by becoming a callback after
we check for the serialization requirement. We can use a default
callback (the current one) for pre-initialization and then then we
can just peg the proper callback upon ath9k_hw_attach_ops(). The
callback for the case where serialization is not required would be
a straight forward read.

This can be done for ath9k_iowrite32() as well.

  Luis
_______________________________________________
ath9k-devel mailing list
ath9k-devel at lists.ath9k.org
https://lists.ath9k.org/mailman/listinfo/ath9k-devel


 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100525/561f1f00/attachment-0001.htm 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: oprofile.AP.log
Url: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100525/561f1f00/attachment-0002.txt 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: oprofile.SU.log
Url: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20100525/561f1f00/attachment-0003.txt