All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sebastian Gottschall <s.gottschall@dd-wrt.com>
To: Koen Vandeputte <koen.vandeputte@citymesh.com>,
	Petrosilius <petrosilius@posteo.de>,
	linux-wireless@vger.kernel.org
Subject: Re: [Bugreport] ath9k dynack not working/low performance on 5 & 10MHz Bandwidth
Date: Wed, 14 Jul 2021 07:38:39 +0200	[thread overview]
Message-ID: <8a147d51-9dc0-5767-5d17-793845b9c5f7@dd-wrt.com> (raw)
In-Reply-To: <a7e37fa2-5fe7-6a39-d72d-9e3723540331@citymesh.com>


>
>
> Above log is with an iperf session running in the background. :-s
>
> Will look further into this ..
>
> Regards,
>
> Koen
some cents from me. the chipset has a hw delay/latency which doubles and 
quadroubles depending on the channel width. this special offset needs to 
be considered for ack
timing calculation. especially at very short distances you will run into 
a problem that if this calculation is wrong, the ack timing register 
contains a lower value than the hw delay/latency.
this will result in a effective negative ack value. maybe its worth to 
check if the values used for this latency offset are correct?
i found also a patch in my own code which i made many years ago (i 
cannot remember why) but maybe it helps
@@ -1128,11 +1110,10 @@

                 slottime = ah->slottime;
         }
+
         /* As defined by IEEE 802.11-2007 17.3.8.6 */
         slottime += 3 * ah->coverage_class;
         acktimeout = slottime + sifstime + ack_offset;
+       eifs += (6 * ah->coverage_class);
         ctstimeout = acktimeout;

         /*

Sebastian

>
>

  reply	other threads:[~2021-07-14  5:39 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-18 11:13 [Bugreport] ath9k dynack not working/low performance on 5 & 10MHz Bandwidth Petrosilius
2021-06-22  9:54 ` Koen Vandeputte
2021-06-22 10:12   ` Petrosilius
2021-06-22 10:40     ` Sebastian Gottschall
2021-06-22 11:54       ` Koen Vandeputte
2021-07-08 22:02       ` Peter Seiderer
2021-06-22 11:52     ` Koen Vandeputte
2021-06-22 11:53       ` Petrosilius
2021-06-22 12:03         ` Koen Vandeputte
2021-06-22 18:54           ` Petrosilius
2021-06-22 21:01             ` Koen Vandeputte
2021-07-13 14:34             ` Koen Vandeputte
2021-07-14  5:38               ` Sebastian Gottschall [this message]
2021-11-09 11:55               ` petrosilius

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=8a147d51-9dc0-5767-5d17-793845b9c5f7@dd-wrt.com \
    --to=s.gottschall@dd-wrt.com \
    --cc=koen.vandeputte@citymesh.com \
    --cc=linux-wireless@vger.kernel.org \
    --cc=petrosilius@posteo.de \
    /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.