Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 9 Aug 2011 01:03:42 +0400
From:      Lev Serebryakov <lev@serebryakov.spb.ru>
To:        Adrian Chadd <adrian@freebsd.org>
Cc:        freebsd-wireless@freebsd.org
Subject:   Re: Bad host-vased AP performace -- and qouple of questions about FreeBSD WiFi stack tuning
Message-ID:  <693584315.20110809010342@serebryakov.spb.ru>
In-Reply-To: <CAJ-VmonrRTRrvR9PWexxVzRzEGS9AW8%2BtZLZqUN1aZT9Uvoqqg@mail.gmail.com>
References:  <1151152206.20110808220720@serebryakov.spb.ru> <CAJ-VmomMh4uvyJGHHkGsi5_C4p3ZJjRwadmstRVe4BjY76QzoA@mail.gmail.com> <1832976164.20110808222108@serebryakov.spb.ru> <CAJ-VmonrRTRrvR9PWexxVzRzEGS9AW8%2BtZLZqUN1aZT9Uvoqqg@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Hello, Adrian.
You wrote 8 =C1=D7=C7=D5=D3=D4=C1 2011 =C7., 22:26:54:

> That's definitely required knowledge. 8.0-RELEASE is a good starting poin=
t.
> I'm afraid it's going to be difficult for me to narrow down what's
> changed without you doing a bit of legwork to find which kernel
> version(s) introduced the performance regression.
> Once we narrow that down, I'll fix it. :)
> Please keep everything else the same though, including the laptop. :)
  It seems, that 8.0-RELEASE behaves slightly different from very
fresh 8-STABLE, but I can not say that it is much better. It is output
from `list sta' when  notebook (client) is placed exactly the same
point as previous experiment:

00:18:de:08:e8:1d    1    9  36M 17.5    0  54585  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 16.0    0  55126  46112 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 17.5    0  55648  50832 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 16.0    0  56169  55632 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 16.0    0  56615  59664 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  24M 16.0    0  57195  65072 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  24M 17.5    0  57936  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  24M 17.0    0  58178  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  24M 16.0    0  58903  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  24M 16.5    0  59465  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  54M 17.5    0  59954  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  54M 17.5    0  60591  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 16.0    0  61343  41344 EPS  AQE     RSN W=
ME

  It looks like it doesn't drop under 24M now. BUT! When I move
notebook into direction of AP (into other room, which is near AP) I've
got:

00:18:de:08:e8:1d    1    9  12M 29.0    0  21255  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  12M 28.0    0  22031  46560 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  12M 30.5    0  22291  46576 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  12M 30.5    0  22291  46576 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  12M 30.5    0  22291  46576 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  12M 30.5    0  22291  46576 EPS  AQEP    RSN W=
ME

  And after door is closed:

00:18:de:08:e8:1d    1    9  12M 23.0    0  27276  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  12M 22.0    0  27789  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  11M 22.5    0  28134  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  11M 22.5    0  28730  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  11M 21.5    0  29040  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  11M 21.5    0  29369  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  12M 22.5    0  29822  41344 EPS  AQE     RSN W=
ME
00:18:de:08:e8:1d    1    9  11M 21.0    0  30232  41344 EPS  AQE     RSN W=
ME

  Better RSSI but much worse speed?!

  I don't have second notebook on hands now to monitor does speed hops
 up and down, but ``ifconfig wlan0 list sta'' every second gives all
 this output.

  It looks like it never works at full and stable 54M now, even if notevook=
 in
direct sight to AP within ~1 meter:

00:18:de:08:e8:1d    1    9  36M 37.5    0  44443  53936 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 37.5    0  44443  53936 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 37.5    0  44443  53936 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 37.5    0  44443  53936 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 37.5    0  44443  53952 EPS  AQEP    RSN W=
ME
00:18:de:08:e8:1d    1    9  36M 37.5    0  44443  53984 EPS  AQEP    RSN W=
ME

 And it seems, that in any other situation it works worse, that it
theoretically can.

 Unfortunately, mini-PCI WiFi cards is not easy to get (and almost
impossible to borrow for free, even for short time) here, and I can
not replace this card in AP and try with another one :(

 I'll try to build 7-STABLE system, but it will take more time, as I
need to adapt configs for it.

--=20
// Black Lion AKA Lev Serebryakov <lev@serebryakov.spb.ru>




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?693584315.20110809010342>