[ath9k-devel] The operation of Minstrel_ht

[ath9k-devel] The operation of Minstrel_ht

[Ali Abedi]

Hello,


I have a few questions about the operation of Minstrel_ht. I couldn't 
find a good reference that
explains minstrel_ht. I tried to understand it by reading the code. The 
fact that frame aggregation
was added to the picture confuses me a bit. I understand the multirate 
retry chain in Minstrel (802.11g).

1- Since the hardware only reties the frame when no block ack is 
received (phy header damage)
(is this correct?), do we still use the multirate retry chain with frame 
aggregation?

2- Is it possible to disable hardware retries?

3- Is software retry defined for an individual MPDU? Is this the number 
of times an MPDU
is rescheduled for transmission?


Thank you in advance.

Best,
Ali

[ath9k-devel] The operation of Minstrel_ht

[Kevin Hayes]

1.  multirate retry is still available for HT aggregates yes.
2.  You would just set the number of retries to 0.  Then if the aggregate
failed (either partially or completely) it would be up to SW to do
something.
3.  Software retry is not defined by the spec, but rather by implementation.
  Retries of MPDU's are done by SW in every case except for the case where
the BA never came back, as you said.



On Fri, Oct 10, 2014 at 7:40 AM, Ali Abedi <a2abedi@uwaterloo.ca> wrote:

> Hello,
>
>
> I have a few questions about the operation of Minstrel_ht. I couldn't
> find a good reference that
> explains minstrel_ht. I tried to understand it by reading the code. The
> fact that frame aggregation
> was added to the picture confuses me a bit. I understand the multirate
> retry chain in Minstrel (802.11g).
>
> 1- Since the hardware only reties the frame when no block ack is
> received (phy header damage)
> (is this correct?), do we still use the multirate retry chain with frame
> aggregation?
>
> 2- Is it possible to disable hardware retries?
>
> 3- Is software retry defined for an individual MPDU? Is this the number
> of times an MPDU
> is rescheduled for transmission?
>
>
> Thank you in advance.
>
> Best,
> Ali
>
>
> _______________________________________________
> ath9k-devel mailing list
> ath9k-devel at lists.ath9k.org
> https://lists.ath9k.org/mailman/listinfo/ath9k-devel
>



-- 
Kevin Hayes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20141031/6b4ecf83/attachment.htm 

[ath9k-devel] The operation of Minstrel_ht

[Ali Abedi]

Thank you for your reply.

Regarding number 1, does  MRR used when no blockack is received?
Because I believe transmission is done when the blockack is received
even if most of the MPDUs are lost.

Thanks,
Ali


On 14-10-31 08:04 PM, Kevin Hayes wrote:
> 1.  multirate retry is still available for HT aggregates yes.
> 2.  You would just set the number of retries to 0.  Then if the 
> aggregate failed (either partially or completely) it would be up to SW 
> to do something.
> 3.  Software retry is not defined by the spec, but rather by 
> implementation.
>   Retries of MPDU's are done by SW in every case except for the case 
> where the BA never came back, as you said.
>
>
>
> On Fri, Oct 10, 2014 at 7:40 AM, Ali Abedi <a2abedi@uwaterloo.ca 
> <mailto:a2abedi@uwaterloo.ca>> wrote:
>
>     Hello,
>
>
>     I have a few questions about the operation of Minstrel_ht. I couldn't
>     find a good reference that
>     explains minstrel_ht. I tried to understand it by reading the
>     code. The
>     fact that frame aggregation
>     was added to the picture confuses me a bit. I understand the multirate
>     retry chain in Minstrel (802.11g).
>
>     1- Since the hardware only reties the frame when no block ack is
>     received (phy header damage)
>     (is this correct?), do we still use the multirate retry chain with
>     frame
>     aggregation?
>
>     2- Is it possible to disable hardware retries?
>
>     3- Is software retry defined for an individual MPDU? Is this the
>     number
>     of times an MPDU
>     is rescheduled for transmission?
>
>
>     Thank you in advance.
>
>     Best,
>     Ali
>
>
>     _______________________________________________
>     ath9k-devel mailing list
>     ath9k-devel at lists.ath9k.org <mailto:ath9k-devel@lists.ath9k.org>
>     https://lists.ath9k.org/mailman/listinfo/ath9k-devel
>
>
>
>
> -- 
> Kevin Hayes

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20141103/38dde532/attachment.htm 

[ath9k-devel] The operation of Minstrel_ht

[Kevin Hayes]

Yes MRR would be used if no BA shows up.
MRR is not used if a BA is received, but SW can lower the rate if it
chooses.

    K++

On Monday, November 3, 2014, Ali Abedi <a2abedi@uwaterloo.ca> wrote:

>  Thank you for your reply.
>
> Regarding number 1, does  MRR used when no blockack is received?
> Because I believe transmission is done when the blockack is received
> even if most of the MPDUs are lost.
>
> Thanks,
> Ali
>
>
> On 14-10-31 08:04 PM, Kevin Hayes wrote:
>
> 1.  multirate retry is still available for HT aggregates yes.
> 2.  You would just set the number of retries to 0.  Then if the aggregate
> failed (either partially or completely) it would be up to SW to do
> something.
> 3.  Software retry is not defined by the spec, but rather by
> implementation.
>   Retries of MPDU's are done by SW in every case except for the case where
> the BA never came back, as you said.
>
>
>
> On Fri, Oct 10, 2014 at 7:40 AM, Ali Abedi <a2abedi@uwaterloo.ca
> <javascript:_e(%7B%7D,'cvml','a2abedi@uwaterloo.ca');>> wrote:
>
>> Hello,
>>
>>
>> I have a few questions about the operation of Minstrel_ht. I couldn't
>> find a good reference that
>> explains minstrel_ht. I tried to understand it by reading the code. The
>> fact that frame aggregation
>> was added to the picture confuses me a bit. I understand the multirate
>> retry chain in Minstrel (802.11g).
>>
>> 1- Since the hardware only reties the frame when no block ack is
>> received (phy header damage)
>> (is this correct?), do we still use the multirate retry chain with frame
>> aggregation?
>>
>> 2- Is it possible to disable hardware retries?
>>
>> 3- Is software retry defined for an individual MPDU? Is this the number
>> of times an MPDU
>> is rescheduled for transmission?
>>
>>
>> Thank you in advance.
>>
>> Best,
>> Ali
>>
>>
>> _______________________________________________
>> ath9k-devel mailing list
>> ath9k-devel at lists.ath9k.org
>> <javascript:_e(%7B%7D,'cvml','ath9k-devel@lists.ath9k.org');>
>> https://lists.ath9k.org/mailman/listinfo/ath9k-devel
>>
>
>
>
>  --
> Kevin Hayes
>
>
>

-- 
Kevin Hayes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ath9k.org/pipermail/ath9k-devel/attachments/20141103/be955f61/attachment.htm 

[ath9k-devel] The operation of Minstrel_ht

[Bonzar]

E m:
V tv

Ali Abedi <a2abedi@uwaterloo.ca> wrote:

>_______________________________________________
>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/20141104/4c0db455/attachment.htm