Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 20 May 2013 12:46:16 +0400
From:      Lev Serebryakov <lev@FreeBSD.org>
To:        Adrian Chadd <adrian@freebsd.org>
Cc:        freebsd-wireless@freebsd.org
Subject:   Re: [rft] please test -HEAD ath; lots of TX changes
Message-ID:  <456290883.20130520124616@serebryakov.spb.ru>
In-Reply-To: <CAJ-Vmokhtz_K7ne8EukbZHQa9QLY3ughYTkHmmqc-ktpov6Uuw@mail.gmail.com>
References:  <CAJ-VmongHgfCBPFpnMLi_5ppVbz%2BtOXVSOzRmCF=yL8FtnMBHQ@mail.gmail.com> <372806514.20130519141024@serebryakov.spb.ru> <CAJ-Vmo=Ak=Af7mdc2LP0niSrGbHsVAFOAen16wCLaEXrj2A82A@mail.gmail.com> <1106213329.20130519193856@serebryakov.spb.ru> <CAJ-VmonobPnfd4u9QnA%2BaCWDWJZYxWFV_N4PwusLyZxiK%2BuHoA@mail.gmail.com> <1377052407.20130519195416@serebryakov.spb.ru> <CAJ-Vmon3UXXEf-fJ6Cc29qRF1U31UJegP-fR3NQ5_zGvv1tRsg@mail.gmail.com> <5931014.20130519213437@serebryakov.spb.ru> <CAJ-VmokuwG3UbCZscR9rTKWtah0gEM4rd0L6kHM6jcx_tCaioQ@mail.gmail.com> <1596494929.20130519223744@serebryakov.spb.ru> <CAJ-Vmokhtz_K7ne8EukbZHQa9QLY3ughYTkHmmqc-ktpov6Uuw@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Hello, Adrian.
You wrote 20 =D0=BC=D0=B0=D1=8F 2013 =D0=B3., 0:27:52:

AC> Right, but you're likely seeing a different issue to "hardware TXQ is s=
tuck."
AC> I think we've solved that. What we're likely seeing now is some odd
AC> buffer exhaustion issues that I haven't yet seen in the driver. -HEAD
AC> is supposed to be limiting how many ath_buf's are queued  per node
AC> _just_ so this particular issue doesn't occur. But it sounds like this
AC> isn't entirely working.
  ...and management frames, which is needed to re-key station failed
  to transmit and station de-associated, do I understand this right?

AC> Now, as for the rate related stuff - yes, MCS15 is actually giving
AC> high (99-100%) success rates. But it may not be transmitting MCS15
AC> very often. How busy is the air? Compile up athsurvey and run that
  Typically air is very busy in my apartments. I'm living in multi-store
multi-apartment building and here are A LOT of private WiFi networks.
Number of them are changing according to time of day (it looks like
many people TURN OFF their routers/APs when they are not at home), but
at night here are 30-35 networks in 2.4Ghz, and at morning, when I run
tests, 15 WiFi networks at range is not unusual.

AC> whilst you're doing a test.
  I'll try it.

AC> I'm going to tidy up the transmit path a little more before I start
AC> working on trying to diagnose why you're seeing the behaviour you are.
AC> But we'll get to it, I promise.
  :) Ok. I want notice, that sometimes (typically AFTER problems with
 de/re-association) it pick ups higher rates and then I see
 125-150Mbit/s UDP stream (it is not full 300Mbit/s, but much better).
 So, looks like problem is not only with busy environment.

AC> If you see any other transmit queue stuck issues then please let me
AC> know. I want high confidence that we've solved those before I move
AC> onto the next stuff.
  Ok.

AC> Thanks for all your patience!
  Thanks for all our hard work!

--=20
// Black Lion AKA Lev Serebryakov <lev@FreeBSD.org>




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?456290883.20130520124616>