All of lore.kernel.org
 help / color / mirror / Atom feed
From: Bhavesh Kamani <bhavesh.kamani@cyberoam.com>
To: ath9k-devel@lists.ath9k.org
Subject: [ath9k-devel] ath9k : Not able to set LongRetry and ShortRetry.
Date: Mon, 12 May 2014 22:19:42 +0530	[thread overview]
Message-ID: <5370FBA6.1060304@cyberoam.com> (raw)
In-Reply-To: <CAJ-Vmo=A5KTrLE5=Z4RmsX9AAhSMNgcAqo9HyUjwvy5SPj8qmw@mail.gmail.com>

Hi Adrian,

Thanks a lot for the detailed info. I will try different combination in 
the pointed function and observe it.

Can you/someone help me what is the meaning of below macros?

#define INIT_SH_RETRY 10
#define INIT_LG_RETRY 10
#define INIT_SSH_RETRY 32
#define INIT_SLG_RETRY 32

Setting value 2 for all these macros, increases the packet drop by 
2%(approximately) in lower signal strength(approx -80 db), if I set the 
values to 64 then packet drop is very less.

Thanks,
Bhavesh.


On Monday 12 May 2014 10:00 PM, Adrian Chadd wrote:
> Hi,
>
> ath9k_11n_rate_series has "Tries". That's how many attempts at that
> given rate series entry.
>
> ar9002_set_txdesc() and ar9003_set_txdesc() (where the latter is
> relevant for you, using an ar9380) turn that rate series information
> into hardware specific information.
>
> ath_buf_set_rate() takes care of populating the bf->rates[] array,
> which is an array of ath9k_11n_rate_series entries. (Note - there's
> four of them. The hardware is limited to four.)
>
> Now, those fields in question behave slightly differently to what I
> think you expect:
>
> * Retry limits (D_RETRY_LIMIT)
>      Offset: 0x1080 + (D << 2)
>
>      3:0   - Frame RTS failure limit.  Specifies the number of times a
>              frame's RTS exchange may fail before the current transmission
>              series is terminated.  (A frame's RTS exchange fails if RTS is
>              enabled for the frame, but when the MAC sends the RTS on the
>              air, no CTS is received.)  Resets to 0x4.  Note: A value of 0x0
>              is UNSUPPORTED.
>      7:4   - Reserved
>      13:8  - Station RTS failure limit.  Specifies the number of times a
>              frame's RTS exchange may fail before the CW is reset to CW_MIN.
>              Resets to 0x20.  Note: A value of 0x0 is UNSUPPORTED.
>      19:14 - Station data failure limit.  Specifies the number of times a
>              frame's data exchange may fail before CW is reset to CW_MIN.
>              Resets to 0x20.  Note: A value of 0x0 is UNSUPPORTED.
>      31:20 - Reserved
>
> .. so frame RTS controls the RTS failure count; I think the rate
> series tries is the total attempts made to transmit the frame assuming
> RTS succeeds, and station RTS/station data have to do with contention
> window management.
>
> Hope that helps,
>
>
>
> -a
>
>
> On 12 May 2014 08:49, Bhavesh Kamani <bhavesh.kamani@cyberoam.com> wrote:
>> Hi Adrian,
>>
>> Can you please point me the exact file/function for rate schedule?
>>
>> But still I am not able to understand the usage of these macros.
>>
>> Thanks,
>> Bhavesh.
>>
>>
>> On Monday 12 May 2014 09:13 PM, Adrian Chadd wrote:
>>> There's also a maximum transmit retry counter in the rate schedule
>>> setup in the transmit descriptor.
>>>
>>> -a
>>>
>>>
>>> On 12 May 2014 07:59, Bhavesh Kamani <bhavesh.kamani@cyberoam.com> wrote:
>>>> Hi Team,
>>>>
>>>> I am using this Ar9380 NIC as AP using hostapd 2.0.
>>>>
>>>> Thanks,
>>>> Bhavesh.
>>>>
>>>> On Monday 12 May 2014 08:26 PM, Bhavesh Kamani wrote:
>>>>> Hi Team,
>>>>>
>>>>> I have changed the values of following four macros and set it to 2 to
>>>>> change LongRetryLimit and ShortRetryLimit, but I am not able to see
>>>>> any difference.
>>>>>
>>>>> File : drivers/net/wireless/ath/ath9k/mac.h
>>>>> #define INIT_SH_RETRY   10
>>>>> #define INIT_LG_RETRY   10
>>>>> #define INIT_SSH_RETRY  32
>>>>> #define INIT_SLG_RETRY  32
>>>>>
>>>>> After setting value 2, I am able to see more than 2 retries when ACK
>>>>> is not sent by WiFi client for particular packet.
>>>>> Please let me know how to change LongRetryLimit and ShortRetryLimit.
>>>>> Is it the right place where I am changing the values? If yes, then
>>>>> what are the possible value ranges for LongRetryLimit and
>>>>> ShortRetryLimit?
>>>>>
>>>>> ath9k version : backports-3.12.8-1
>>>>> Hardware : AR9380 miniPCIe
>>>>>
>>>>> Thanks,
>>>>> Bhavesh.
>>>> _______________________________________________
>>>> ath9k-devel mailing list
>>>> ath9k-devel at lists.ath9k.org
>>>> https://lists.ath9k.org/mailman/listinfo/ath9k-devel
>>

  reply	other threads:[~2014-05-12 16:49 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-05-12 14:56 [ath9k-devel] ath9k : Not able to set LongRetry and ShortRetry Bhavesh Kamani
2014-05-12 14:59 ` Bhavesh Kamani
2014-05-12 15:43   ` Adrian Chadd
2014-05-12 15:49     ` Bhavesh Kamani
2014-05-12 16:30       ` Adrian Chadd
2014-05-12 16:49         ` Bhavesh Kamani [this message]
2014-05-12 19:11           ` Adrian Chadd
2014-05-14 13:13         ` Bhavesh Kamani
2014-05-14 20:15           ` Adrian Chadd

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5370FBA6.1060304@cyberoam.com \
    --to=bhavesh.kamani@cyberoam.com \
    --cc=ath9k-devel@lists.ath9k.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.