Hi guys,
More experimentation with my rtl8187. With some playing around with the
access point location, I think I may have stumbled on a good position.
Anyway ...
I noticed that the rate reported by iwconfig only ever gets to 11Mb/s
despite the fact that I now have a signal level of 90/100 - 100/100. I
found that I can do
iwconfig wlan0 rate 54M
but it seems that although the card is still associated networking does=
nt
work (pings fail). If I change it back to auto
iwconfig wlan0 rate auto
then iwconfig reports 11Mb/s again and pinging resumes fine.
Is there any chance that the driver is incorrectly reporting its mode o=
r
something in that the rates it displays bear the hallmarks of a b card
rather than g?
Using the compat-wireless snapshot from 9/2/2009
Mandriva 2.6.27.10-desktop586-1mnb
Any ideas appreciated.
Regards
PGR
--=20
Laws are partly formed for the sake of good men, in order to instruct t=
hem
how they may live on friendly terms with one another, and partly for th=
e
sake of those who refuse to be instructed, whose spirit cannot be subdu=
ed,
or softened, or hindered from plunging into evil. [The Laws, Plato]
You fiend! Never have I encountered such corrupt and foul-minded
perversity=85 Have you ever considered a career in the church?
Bishop of Bath and Wells (Blackadder II)
> P.G. Richardson wrote:
>> Hi guys,
>>
>> More experimentation with my rtl8187. With some playing around with =
the
>> access point location, I think I may have stumbled on a good positio=
n.
>>
>> Anyway ...
>>
>> I noticed that the rate reported by iwconfig only ever gets to 11Mb/=
s
>> despite the fact that I now have a signal level of 90/100 - 100/100.=
I
>> found that I can do
>>
>> iwconfig wlan0 rate 54M
>>
>> but it seems that although the card is still associated networking
>> doesnt
>> work (pings fail). If I change it back to auto
>>
>> iwconfig wlan0 rate auto
>>
>> then iwconfig reports 11Mb/s again and pinging resumes fine.
>>
>> Is there any chance that the driver is incorrectly reporting its mod=
e or
>> something in that the rates it displays bear the hallmarks of a b ca=
rd
>> rather than g?
>>
>> Using the compat-wireless snapshot from 9/2/2009
>> Mandriva 2.6.27.10-desktop586-1mnb
>
> What rate-adjustment method are you using? The output of dmesg will t=
ell.
> If you
> are using minstrel, then you should post the contents from the follow=
ing
> commands:
>
> STATS=3D$(find /sys -name rc_stats)
> echo $STATS
> sudo cat $STATS
>
> Larry
>
phy0: Selected rate control algorithm 'minstrel'
find /sys -name rc_stats produces no results whatsoever. These is no
rc_stats in /sys.
Regards
PGR
--=20
Laws are partly formed for the sake of good men, in order to instruct t=
hem
how they may live on friendly terms with one another, and partly for th=
e
sake of those who refuse to be instructed, whose spirit cannot be subdu=
ed,
or softened, or hindered from plunging into evil. [The Laws, Plato]
You fiend! Never have I encountered such corrupt and foul-minded
perversity=85 Have you ever considered a career in the church?
Bishop of Bath and Wells (Blackadder II)
P.G. Richardson wrote:
> Hi Larry,
>
> Been experimenting over last couple of days to enable this. Have
> successfully recompiled the kernel and installed with above config option
> enabled. This may have been unnecessary but could not compile without
> doing it.
>
> Anyway, now running the custom kernel and installed compart-wireless
> snapshot from 09/02/2009 and still nothing has appeared using
> find /sys -name rc_stats
>
> I must be missing some vital step. Any thoughts please.
The configuration needs to be in compat-wireless. Please apply the following
patch and rebuild the compat-wireless package.
Index: compat-wireless-2009-02-19/config.mk
===================================================================
--- compat-wireless-2009-02-19.orig/config.mk
+++ compat-wireless-2009-02-19/config.mk
@@ -74,6 +74,7 @@ endif # kernel Makefile check
# Wireless subsystem stuff
CONFIG_MAC80211=m
+CONFIG_MAC80211_DEBUGFS=y
# choose between pid and minstrel as default rate control algorithm
CONFIG_MAC80211_RC_DEFAULT=minstrel
Larry
Right, in business ... took an effort but there we are.
Still couldnt get any info from "find /sys -name rc_stats". However, I
found an old thread from last year which interestingly seems exactly th=
e
problem.
http://www.gossamer-threads.com/lists/linux/kernel/860544
So have done similar and mounted debugfs at /debug. This is the results=
of
cat /debug/ieee80211/phy0/stations/f6cefa14/rc_stats:
rate throughput ewma prob this prob this succ/attempt success=
=20
attempts
tP 1 0.8 89.5 100.0 1( 1) 102 =
=20
176
2 0.7 39.3 50.0 0( 0) 11 =
=20
56
T 5.5 3.4 68.7 50.0 1( 2) 1065 =
=20
3462
11 0.2 2.4 0.0 0( 0) 1158 =
=20
1463
6 0.1 2.6 0.0 0( 0) 232 =
=20
935
9 0.0 0.7 0.0 0( 0) 17 =
=20
159
12 0.0 0.7 0.0 0( 0) 106 =
=20
338
18 0.0 0.4 0.0 0( 0) 52 =
=20
246
24 0.0 0.1 0.0 0( 0) 17 =
=20
207
36 0.0 0.1 0.0 0( 0) 10 =
=20
143
48 0.0 0.0 0.0 0( 0) 0 =
=20
28
54 0.0 0.0 0.0 0( 0) 0 =
=20
28
Total packet count:: ideal 2863 lookaround 150
Let me know if this means anything or you need further info. Certainly,
seems to confirm that connection never seems to go above 11Mb/s.
Cheers
PGR
--=20
Laws are partly formed for the sake of good men, in order to instruct t=
hem
how they may live on friendly terms with one another, and partly for th=
e
sake of those who refuse to be instructed, whose spirit cannot be subdu=
ed,
or softened, or hindered from plunging into evil. [The Laws, Plato]
You fiend! Never have I encountered such corrupt and foul-minded
perversity=85 Have you ever considered a career in the church?
Bishop of Bath and Wells (Blackadder II)
> P.G. Richardson wrote:
>>> P.G. Richardson wrote:
>>>> Hi guys,
>>>>
>>>> More experimentation with my rtl8187. With some playing around wit=
h
>>>> the
>>>> access point location, I think I may have stumbled on a good posit=
ion.
>>>>
>>>> Anyway ...
>>>>
>>>> I noticed that the rate reported by iwconfig only ever gets to 11M=
b/s
>>>> despite the fact that I now have a signal level of 90/100 - 100/10=
0. I
>>>> found that I can do
>>>>
>>>> iwconfig wlan0 rate 54M
>>>>
>>>> but it seems that although the card is still associated networking
>>>> doesnt
>>>> work (pings fail). If I change it back to auto
>>>>
>>>> iwconfig wlan0 rate auto
>>>>
>>>> then iwconfig reports 11Mb/s again and pinging resumes fine.
>>>>
>>>> Is there any chance that the driver is incorrectly reporting its m=
ode
>>>> or
>>>> something in that the rates it displays bear the hallmarks of a b =
card
>>>> rather than g?
>>>>
>>>> Using the compat-wireless snapshot from 9/2/2009
>>>> Mandriva 2.6.27.10-desktop586-1mnb
>>> What rate-adjustment method are you using? The output of dmesg will
>>> tell.
>>> If you
>>> are using minstrel, then you should post the contents from the
>>> following
>>> commands:
>>>
>>> STATS=3D$(find /sys -name rc_stats)
>>> echo $STATS
>>> sudo cat $STATS
>>>
>>> Larry
>>>
>>
>> phy0: Selected rate control algorithm 'minstrel'
>>
>> find /sys -name rc_stats produces no results whatsoever. These is no
>> rc_stats in /sys.
>
> You need to set "CONFIG_MAC80211_DEBUGFS=3Dy" in your configuration.
>
Hi Larry,
Been experimenting over last couple of days to enable this. Have
successfully recompiled the kernel and installed with above config opti=
on
enabled. This may have been unnecessary but could not compile without
doing it.
Anyway, now running the custom kernel and installed compart-wireless
snapshot from 09/02/2009 and still nothing has appeared using
find /sys -name rc_stats
I must be missing some vital step. Any thoughts please.
Cheers
PGR
--=20
Laws are partly formed for the sake of good men, in order to instruct t=
hem
how they may live on friendly terms with one another, and partly for th=
e
sake of those who refuse to be instructed, whose spirit cannot be subdu=
ed,
or softened, or hindered from plunging into evil. [The Laws, Plato]
You fiend! Never have I encountered such corrupt and foul-minded
perversity=85 Have you ever considered a career in the church?
Bishop of Bath and Wells (Blackadder II)
> I'm not an expert in reading these tables, but the success rates abo=
ve 11
> Mb/s
> are very low, which seems to indicate that 11 Mb/s is a reasonable ra=
te
> for your
> device. There is, however, a possibility that the code is not reporti=
ng
> the
> retry count correctly. I know from your initial post that 54M doesn't
> work, but
> can you get any throughput at 24, 36, or 48M? There is also the
> possibility that
> your hardware is broken for OFDM encoding. That scheme is used for 6,=
9,
> and all
> rates 18M and above.
Minimal throughput for fixed rate of 6M. Pinging only replies to every =
4th
packet or so. No ping replies at all for 9, 18, 24, 36, 48 or 54.
Not sure what the actual rate of packet delivery is (doesnt seem slower
than my other laptop) but can confirm with iperf later.
Could also dump an XP installation on here to confirm whether this is a=
n
hardware issue.
Regards
PGR
--=20
Laws are partly formed for the sake of good men, in order to instruct t=
hem
how they may live on friendly terms with one another, and partly for th=
e
sake of those who refuse to be instructed, whose spirit cannot be subdu=
ed,
or softened, or hindered from plunging into evil. [The Laws, Plato]
You fiend! Never have I encountered such corrupt and foul-minded
perversity=85 Have you ever considered a career in the church?
Bishop of Bath and Wells (Blackadder II)
P.G. Richardson wrote:
> Hi guys,
>
> More experimentation with my rtl8187. With some playing around with the
> access point location, I think I may have stumbled on a good position.
>
> Anyway ...
>
> I noticed that the rate reported by iwconfig only ever gets to 11Mb/s
> despite the fact that I now have a signal level of 90/100 - 100/100. I
> found that I can do
>
> iwconfig wlan0 rate 54M
>
> but it seems that although the card is still associated networking doesnt
> work (pings fail). If I change it back to auto
>
> iwconfig wlan0 rate auto
>
> then iwconfig reports 11Mb/s again and pinging resumes fine.
>
> Is there any chance that the driver is incorrectly reporting its mode or
> something in that the rates it displays bear the hallmarks of a b card
> rather than g?
>
> Using the compat-wireless snapshot from 9/2/2009
> Mandriva 2.6.27.10-desktop586-1mnb
What rate-adjustment method are you using? The output of dmesg will tell. If you
are using minstrel, then you should post the contents from the following commands:
STATS=$(find /sys -name rc_stats)
echo $STATS
sudo cat $STATS
Larry
P.G. Richardson wrote:
> Right, in business ... took an effort but there we are.
>
> Still couldnt get any info from "find /sys -name rc_stats". However, I
> found an old thread from last year which interestingly seems exactly the
> problem.
>
> http://www.gossamer-threads.com/lists/linux/kernel/860544
>
> So have done similar and mounted debugfs at /debug. This is the results of
> cat /debug/ieee80211/phy0/stations/f6cefa14/rc_stats:
My /etc/fsconfig has these two lines:
sysfs /sys sysfs noauto 0 0
debugfs /sys/kernel/debug debugfs noauto 0 0
For getting help with issues like this, you should probably mount debugfs this
way as well. AFAIK, these were set up by openSUSE.
> rate throughput ewma prob this prob this succ/attempt success
> attempts
> tP 1 0.8 89.5 100.0 1( 1) 102
> 176
> 2 0.7 39.3 50.0 0( 0) 11
> 56
> T 5.5 3.4 68.7 50.0 1( 2) 1065
> 3462
> 11 0.2 2.4 0.0 0( 0) 1158
> 1463
> 6 0.1 2.6 0.0 0( 0) 232
> 935
> 9 0.0 0.7 0.0 0( 0) 17
> 159
> 12 0.0 0.7 0.0 0( 0) 106
> 338
> 18 0.0 0.4 0.0 0( 0) 52
> 246
> 24 0.0 0.1 0.0 0( 0) 17
> 207
> 36 0.0 0.1 0.0 0( 0) 10
> 143
> 48 0.0 0.0 0.0 0( 0) 0
> 28
> 54 0.0 0.0 0.0 0( 0) 0
> 28
>
> Total packet count:: ideal 2863 lookaround 150
>
> Let me know if this means anything or you need further info. Certainly,
> seems to confirm that connection never seems to go above 11Mb/s.
I'm not an expert in reading these tables, but the success rates above 11 Mb/s
are very low, which seems to indicate that 11 Mb/s is a reasonable rate for your
device. There is, however, a possibility that the code is not reporting the
retry count correctly. I know from your initial post that 54M doesn't work, but
can you get any throughput at 24, 36, or 48M? There is also the possibility that
your hardware is broken for OFDM encoding. That scheme is used for 6, 9, and all
rates 18M and above.
Larry
P.G. Richardson wrote:
>> P.G. Richardson wrote:
>>> Hi guys,
>>>
>>> More experimentation with my rtl8187. With some playing around with the
>>> access point location, I think I may have stumbled on a good position.
>>>
>>> Anyway ...
>>>
>>> I noticed that the rate reported by iwconfig only ever gets to 11Mb/s
>>> despite the fact that I now have a signal level of 90/100 - 100/100. I
>>> found that I can do
>>>
>>> iwconfig wlan0 rate 54M
>>>
>>> but it seems that although the card is still associated networking
>>> doesnt
>>> work (pings fail). If I change it back to auto
>>>
>>> iwconfig wlan0 rate auto
>>>
>>> then iwconfig reports 11Mb/s again and pinging resumes fine.
>>>
>>> Is there any chance that the driver is incorrectly reporting its mode or
>>> something in that the rates it displays bear the hallmarks of a b card
>>> rather than g?
>>>
>>> Using the compat-wireless snapshot from 9/2/2009
>>> Mandriva 2.6.27.10-desktop586-1mnb
>> What rate-adjustment method are you using? The output of dmesg will tell.
>> If you
>> are using minstrel, then you should post the contents from the following
>> commands:
>>
>> STATS=$(find /sys -name rc_stats)
>> echo $STATS
>> sudo cat $STATS
>>
>> Larry
>>
>
> phy0: Selected rate control algorithm 'minstrel'
>
> find /sys -name rc_stats produces no results whatsoever. These is no
> rc_stats in /sys.
You need to set "CONFIG_MAC80211_DEBUGFS=y" in your configuration.
Larry
Hi Larry,
>> Is there anything I can do to help improve this situation? These res=
ults
>> seems to rule out hardware in favour of being driver related. Are th=
ere
>> other tests I could perform to determine where the fragility lies?
>>
>> Any thoughts, let me know.
>
> OK, the Windows driver is better than the Linux one. Unfortunately, s=
ource
> is not available.
Unsurprisingly, I imagine that the windows environment was what the
manufacturer had in mind. I mean no disrespect to anyone regarding thes=
e
tests. They simply serve to rule out hardware failures and mean that I =
can
help in my probably small and naive way to improve the driver.
> You can check what we did when the RTL8187B was implemented. To do th=
at,
> you
> need to download the Realtek vendor-written Linux driver and check wh=
at
> they do
> against what is in the current in-kernel driver. The critical areas w=
ill
> likely
> be associated with power settings for the radio. Be aware that their =
code
> is
> quite difficult to read as it has multiple conditional compilations t=
hat
> are
> hard to keep straight.
Okay, I shall see how I go. I doubt my ability to come up with somethin=
g
awe inspiring but if I can help and contribute then I shall make it a
point.
> I will be happy to test any improvements that you suggest.
Thanks
PGR
--=20
Laws are partly formed for the sake of good men, in order to instruct t=
hem
how they may live on friendly terms with one another, and partly for th=
e
sake of those who refuse to be instructed, whose spirit cannot be subdu=
ed,
or softened, or hindered from plunging into evil. [The Laws, Plato]
You fiend! Never have I encountered such corrupt and foul-minded
perversity=85 Have you ever considered a career in the church?
Bishop of Bath and Wells (Blackadder II)
P.G. Richardson wrote:
> Is there anything I can do to help improve this situation? These results
> seems to rule out hardware in favour of being driver related. Are there
> other tests I could perform to determine where the fragility lies?
>
> Any thoughts, let me know.
OK, the Windows driver is better than the Linux one. Unfortunately, source is
not available.
You can check what we did when the RTL8187B was implemented. To do that, you
need to download the Realtek vendor-written Linux driver and check what they do
against what is in the current in-kernel driver. The critical areas will likely
be associated with power settings for the radio. Be aware that their code is
quite difficult to read as it has multiple conditional compilations that are
hard to keep straight.
I will be happy to test any improvements that you suggest.
Larry
> I'm not an expert in reading these tables, but the success rates abov=
e 11
> Mb/s
> are very low, which seems to indicate that 11 Mb/s is a reasonable ra=
te
> for your
> device. There is, however, a possibility that the code is not reporti=
ng
> the
> retry count correctly. I know from your initial post that 54M doesn't
> work, but
> can you get any throughput at 24, 36, or 48M? There is also the
> possibility that
> your hardware is broken for OFDM encoding. That scheme is used for 6,=
9,
> and all
> rates 18M and above.
>
> Larry
=46urther playing around with my rtl8187b card. I have installed window=
s
onto my laptop too for performance comparisons. Just how crude or accur=
ate
such comparisons are I dont know so if anyone has any experience then d=
o
shout.
After installing windows, I did the following:
1) ping -n 10000 [server]
Result: Ping replies were good from throughout the house. Only 1 packet
lost out of about 600.
2) Observation of wireless connection
Result: From the little diagnostics there are on windows, it reckons it
connected at 54g (completely false as windows might just say that anywa=
y).
Walking around the house, the signal bars only ever went down to 3 out =
of
5.
3) iperf -c [server]
Result: From 3 different locations in the house, the performance is
reported at 5.5 - 6Mbits/sec.
Rebooted back into Mandriva 2009 using the compat-wireless snapshot fro=
m
the 24/2/2009.
1) ping [server]
Result: Ping is brittle. Movements around the house can ensure a packet=
is
lost with minimal effort. Locations away from the router but within ran=
ge,
that could be pinged from under Windows, resulted in lost packets and e=
ven
"Destination unreachables". Slight turnings of the laptop either left o=
r
right when placed on a table impact ping connectivity.
2) Observation of wireless connection
Result: running iwconfig and looking at the signal strength indicates a
range of 28-100. The locations where 28 occurs rarely get a decent ping
test.
3) iperf -c [server]
Result: From 3 different locations in the house, the performance is
reported at 3 - 4Mbits/sec.
So my rather crude conclusions at the moment is the rtl8187 under mandr=
iva
performs at a slower rate than under windows. More interestingly, it se=
ems
that its receiver is not performing as well either, meaning the signal
coverage area for my laptop is much smaller.
Is there anything I can do to help improve this situation? These result=
s
seems to rule out hardware in favour of being driver related. Are there
other tests I could perform to determine where the fragility lies?
Any thoughts, let me know.
Regards
PGR
--=20
Laws are partly formed for the sake of good men, in order to instruct t=
hem
how they may live on friendly terms with one another, and partly for th=
e
sake of those who refuse to be instructed, whose spirit cannot be subdu=
ed,
or softened, or hindered from plunging into evil. [The Laws, Plato]
You fiend! Never have I encountered such corrupt and foul-minded
perversity=85 Have you ever considered a career in the church?
Bishop of Bath and Wells (Blackadder II)