[ath9k-devel] Disabling link-layer ACKs

[ath9k-devel] Disabling link-layer ACKs

[William Usher]

Hi everyone.
Back when we used Madwifi for our protocol (hop:
http://lass.cs.umass.edu/~mingli/hop/) we were able to disable link-layer
ACKs, is this possible in the ath9k driver? I've looked through the source,
but nothing popped out at me.

It is my understanding that link layer ACKs are generated in hardware so I'm
not sure how this would be accomplished in a driver without a HAL.

Thanks for your help,
Will
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20080729/c9adf638/attachment.htm 

[ath9k-devel] Disabling Link-Layer Acks

[Matt Tierney]

This is bumping an old question <
http://www.mail-archive.com/ath9k-devel@lists.ath9k.org/msg00016.html> but I
was wondering if there was a simple (quick and dirty, if need be) way to
disable link-layer acks.

I think I found the right place to do this in the xmit.c code, under the
setup_tx_flags function:

*        if (tx_info->flags & IEEE80211_TX_CTL_NO_ACK)
                flags |= ATH9K_TXDESC_NOACK;*

I expect that I would just need to comment out the first line with the
if-condition.

Is this a correct way to handle disabling link-layer acks for now?

Thanks,
Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20091001/25e64a71/attachment.htm 

[ath9k-devel] Disabling Link-Layer Acks

[Luis R. Rodriguez]

On Thu, Oct 01, 2009 at 09:50:47AM -0700, Matt Tierney wrote:
> This is bumping an old question <http://www.mail-archive.com/ath9k-devel@lists.ath9k.org/msg00016.html> but I was wondering if there was a simple (quick and dirty, if need be) way to disable link-layer acks.
> 
> I think I found the right place to do this in the xmit.c code, under the setup_tx_flags function:
> 
>         if (tx_info->flags & IEEE80211_TX_CTL_NO_ACK)
>                 flags |= ATH9K_TXDESC_NOACK;
> 
> I expect that I would just need to comment out the first line with the if-condition.
> 
> Is this a correct way to handle disabling link-layer acks for now?

Not sure, I'd be checking the code just as you are but I will not another
thing you may want to look at also is the extremely well documented
ath_hw_setbssidmask():

http://git.kernel.org/?p=linux/kernel/git/linville/wireless-testing.git;a=blob;f=drivers/net/wireless/ath/hw.c;h=ecc9eb01f4fa12bddfd7ab044c79847bfe643c84;hb=HEAD

  Luis

[ath9k-devel] Disabling Link-Layer Acks

[Sujith]

Matt Tierney wrote:
> This is bumping an old question <http://www.mail-archive.com/ath9k-devel@lists.ath9k.org/msg00016.html> but I was wondering if there was a simple (quick and dirty, if need be) way to disable link-layer acks.
> 
> I think I found the right place to do this in the xmit.c code, under the setup_tx_flags function:
> 
>         if (tx_info->flags & IEEE80211_TX_CTL_NO_ACK)
>                 flags |= ATH9K_TXDESC_NOACK;
> 
> I expect that I would just need to comment out the first line with the if-condition.
> 
> Is this a correct way to handle disabling link-layer acks for now?

No. That flag is used to indicate whether we should
expect an ACK for the frame being sent.

The correct way would be to set AR_DIAG_ACK_DIS in AR_DIAG_SW (reg.h)

Sujith

[ath9k-devel] Disabling Link-Layer Acks

[Valentin Manea]

On 10/07/2009 09:12 AM, Sujith wrote:
> Matt Tierney wrote:
>    
>> This is bumping an old question<http://www.mail-archive.com/ath9k-devel@lists.ath9k.org/msg00016.html>  but I was wondering if there was a simple (quick and dirty, if need be) way to disable link-layer acks.
>>
>> I think I found the right place to do this in the xmit.c code, under the setup_tx_flags function:
>>
>>          if (tx_info->flags&  IEEE80211_TX_CTL_NO_ACK)
>>                  flags |= ATH9K_TXDESC_NOACK;
>>
>> I expect that I would just need to comment out the first line with the if-condition.
>>
>> Is this a correct way to handle disabling link-layer acks for now?
>>      
> No. That flag is used to indicate whether we should
> expect an ACK for the frame being sent.
>
> The correct way would be to set AR_DIAG_ACK_DIS in AR_DIAG_SW (reg.h)
>    
Hi Sujit,

   As I'm interested in this, I was wondering if setting this bit also 
disables sending the BlockAcks?


Thanks

[ath9k-devel] Disabling Link-Layer Acks

[Ranga Rao Ravuri]

You cannot disable ACK-rx, as mentioned here, setting DIAG bit should help 
*not* sending ACK
~ R


On Wed, 7 Oct 2009, Valentin Manea wrote:

> 
> 
> On 10/07/2009 09:12 AM, Sujith wrote:
> > Matt Tierney wrote:
> >
> >> This is bumping an old question<http://www.mail-archive.com/ath9k-devel@lists.ath9k.org/msg00016.html>  but I was wondering if there was a simple (quick and dirty, if need be) way to disable link-layer acks.
> >>
> >> I think I found the right place to do this in the xmit.c code, under the setup_tx_flags function:
> >>
> >>          if (tx_info->flags&  IEEE80211_TX_CTL_NO_ACK)
> >>                  flags |= ATH9K_TXDESC_NOACK;
> >>
> >> I expect that I would just need to comment out the first line with the if-condition.
> >>
> >> Is this a correct way to handle disabling link-layer acks for now?
> >>
> > No. That flag is used to indicate whether we should
> > expect an ACK for the frame being sent.
> >
> > The correct way would be to set AR_DIAG_ACK_DIS in AR_DIAG_SW (reg.h)
> >
> Hi Sujit,
> 
>    As I'm interested in this, I was wondering if setting this bit also
> disables sending the BlockAcks?
> 
> 
> Thanks
> _______________________________________________
> ath9k-devel mailing list
> ath9k-devel at lists.ath9k.org
> https://lists.ath9k.org/mailman/listinfo/ath9k-devel
> 

[ath9k-devel] Disabling Link-Layer Acks

[Valentin Manea]

Thanks for your answer.
I did some tests and setting the diag bit also disables BlockAck being 
sent in HT mode.

Best Regards,
Valentin

On 10/07/2009 08:11 PM, Ranga Rao Ravuri wrote:
> You cannot disable ACK-rx, as mentioned here, setting DIAG bit should help
> *not* sending ACK
> ~ R
>
>
> On Wed, 7 Oct 2009, Valentin Manea wrote:
>
>    
>>
>> On 10/07/2009 09:12 AM, Sujith wrote:
>>      
>>> Matt Tierney wrote:
>>>
>>>        
>>>> This is bumping an old question<http://www.mail-archive.com/ath9k-devel@lists.ath9k.org/msg00016.html>   but I was wondering if there was a simple (quick and dirty, if need be) way to disable link-layer acks.
>>>>
>>>> I think I found the right place to do this in the xmit.c code, under the setup_tx_flags function:
>>>>
>>>>           if (tx_info->flags&   IEEE80211_TX_CTL_NO_ACK)
>>>>                   flags |= ATH9K_TXDESC_NOACK;
>>>>
>>>> I expect that I would just need to comment out the first line with the if-condition.
>>>>
>>>> Is this a correct way to handle disabling link-layer acks for now?
>>>>
>>>>          
>>> No. That flag is used to indicate whether we should
>>> expect an ACK for the frame being sent.
>>>
>>> The correct way would be to set AR_DIAG_ACK_DIS in AR_DIAG_SW (reg.h)
>>>
>>>        
>> Hi Sujit,
>>
>>     As I'm interested in this, I was wondering if setting this bit also
>> disables sending the BlockAcks?
>>
>>
>> Thanks
>> _______________________________________________
>> ath9k-devel mailing list
>> ath9k-devel at lists.ath9k.org
>> https://lists.ath9k.org/mailman/listinfo/ath9k-devel
>>
>>